US20030133510A1 - Image coded data re-encoding apparatus - Google Patents
Image coded data re-encoding apparatus Download PDFInfo
- Publication number
- US20030133510A1 US20030133510A1 US09/250,404 US25040499A US2003133510A1 US 20030133510 A1 US20030133510 A1 US 20030133510A1 US 25040499 A US25040499 A US 25040499A US 2003133510 A1 US2003133510 A1 US 2003133510A1
- Authority
- US
- United States
- Prior art keywords
- coded data
- image coded
- image
- signal processing
- encoding apparatus
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Abandoned
Links
Images
Classifications
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04N—PICTORIAL COMMUNICATION, e.g. TELEVISION
- H04N19/00—Methods or arrangements for coding, decoding, compressing or decompressing digital video signals
- H04N19/40—Methods or arrangements for coding, decoding, compressing or decompressing digital video signals using video transcoding, i.e. partial or full decoding of a coded input stream followed by re-encoding of the decoded output stream
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04N—PICTORIAL COMMUNICATION, e.g. TELEVISION
- H04N19/00—Methods or arrangements for coding, decoding, compressing or decompressing digital video signals
- H04N19/10—Methods or arrangements for coding, decoding, compressing or decompressing digital video signals using adaptive coding
- H04N19/102—Methods or arrangements for coding, decoding, compressing or decompressing digital video signals using adaptive coding characterised by the element, parameter or selection affected or controlled by the adaptive coding
- H04N19/124—Quantisation
- H04N19/126—Details of normalisation or weighting functions, e.g. normalisation matrices or variable uniform quantisers
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04N—PICTORIAL COMMUNICATION, e.g. TELEVISION
- H04N19/00—Methods or arrangements for coding, decoding, compressing or decompressing digital video signals
- H04N19/46—Embedding additional information in the video signal during the compression process
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04N—PICTORIAL COMMUNICATION, e.g. TELEVISION
- H04N19/00—Methods or arrangements for coding, decoding, compressing or decompressing digital video signals
- H04N19/10—Methods or arrangements for coding, decoding, compressing or decompressing digital video signals using adaptive coding
- H04N19/102—Methods or arrangements for coding, decoding, compressing or decompressing digital video signals using adaptive coding characterised by the element, parameter or selection affected or controlled by the adaptive coding
- H04N19/124—Quantisation
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04N—PICTORIAL COMMUNICATION, e.g. TELEVISION
- H04N19/00—Methods or arrangements for coding, decoding, compressing or decompressing digital video signals
- H04N19/60—Methods or arrangements for coding, decoding, compressing or decompressing digital video signals using transform coding
Abstract
An image coded data re-encoding apparatus (30) which generates in an image coded data analyzer (310) coded data after signal processing (221) by performing a first digital signal processing on first image coded data (220), supplies an image coded data synthesizer (320) with the coded data after signal processing (221) and multiple signals (222) associated with the first image coded data, and generates in the image coded data synthesizer (320) a second image coded data (240) by performing on the coded data after signal processing (221) a second digital signal processing based on multiple signals (222). The second image coded data is generated without once decoding the first image coded data. This makes it possible to prevent degradation in image quality, delay involved in the transform and an increase of the device size.
Description
- 1. Field of the Invention
- The present invention relates to an image coded data re-encoding apparatus for producing a second image coded data by applying digital signal processing to a first image coded data which is obtained by coding a digital input image signal.
- 2. Description of Related Art
- FIG. 16 is a block diagram showing a conventional image coded data re-encoding apparatus disclosed in Japanese patent application laid-open No. 2-179186/1990, for example. It is an example of a multiple site video conference system. In this figure, the
reference numeral 1 designates a master station, 2 designates a relay station, 3 designates a slave station, 10 designates an image coder of the stations, and 20 designates an image decoder. Thereference numerals 101 designates an input image, 102 designates image coded data, 103 designates a decoded image signal, 104 designates image re-encoded data, and 105 designates a decoded image signal. The operation will now be described. - The
image coder 10 of themaster station 1 performs coding of theinput image 101, and sends the image codeddata 102 to therelay station 2. Therelay station 2 receives the image codeddata 102 with itsimage decoder 20 to decode it, and generates theimage re-encoded data 104 by re-encoding the decodedimage signal 103 with theimage coder 10. The image re-encodeddata 104 thus generated by the re-encoding is transmitted to theslave station 3. Theslave station 3 decodes with theimage decoder 20 theimage re-encoded data 104 which is relayed through therelay station 2, and uses it as thedecoded image signal 105. - When holding a conference using the
relay station 2 with such a decoding and relaying function, it often occurs that themaster station 1 and theslave station 3 employ different coding systems. In this case, it becomes necessary to change the amount of coded data to be generated and various types of parameters such as image size and a frame rate. Thus, therelay station 2 once decodes the received image codeddata 102 into the decodedimage signal 103, and then re-encodes it into theimage re-encoded data 104, thereby matching the different coding systems. - In this way, the conventional image coded data re-encoding apparatus has a process through which the image coded data is once decoded image to be re-encoded to achieve relaying or copying of the image coded data.
- Since the conventional image coded data re-encoding apparatus with such an arrangement once decodes the image coded
data 102 into the decodedimage signal 103, and then re-encodes the decodedimage signal 103 regardless of its contents to relay or convert the image codeddata 102, it has some problems such as degrading the image quality of the decodedimage signal 105, increasing a delay involved in the relay and transform, and augmenting the size of the apparatus. - The present invention is accomplished to solve such problems, and to provide an image coded data re-encoding apparatus which can reduce the image degradation, shorten the processing delay, and shrink the device size, thereby achieving an efficient transform of the image coded data.
- According to one aspect of the present invention, there is provided an image coded data re-encoding apparatus comprising: an image coded data analyzer for generating coded data after signal processing by performing a first digital signal processing on a first image coded data; and an image coded data synthesizer for generating a second image coded data by performing on the coded data after signal processing a second digital signal processing based on multiple signals associated with a first image coded data by using the coded data after signal processing output from the image coded data analyzer and the multiple signals.
- This will offer an advantage of reducing the degradation of the image quality after the transform, decreasing processing delay, and achieving an image coded data re-encoding apparatus with a reduced size in comparison with a system in which the first image coded data is always once decoded into a decoded image, followed by re-encoding of the decoded image into the second image coded data regardless of the decoded data.
- In image coded data re-encoding apparatus, the image coded data analyzer may extract the multiple signals in the course of generating the coded data after signal processing by performing the first digital signal processing on the first image coded data, and may provides the image coded data synthesizer with the coded data after signal processing.
- This will offer an advantage of providing an information effective image coded data re-encoding apparatus capable of obviating special additional information which was needed for the second digital signal processing for generating the second image coded data.
- In the image coded data re-encoding apparatus, the image coded data re-encoding apparatus may further comprise a separator for separating from the first image coded data the multiple signals which have been externally combined with the first image coded data and cannot be extracted from the first image coded data in the first digital signal processing for generating the coded data after signal processing, and the image coded data synthesizer may generate the second image coded data by using the multiple signals output from the separator.
- This will offer an advantage of providing a coded data re-encoding apparatus which can use, in the second digital signal processing for generating the second image coded data, information which cannot be extracted in course of the first digital signal processing, and this will make possible more effective transform than when such information is not used.
- The image coded data re-encoding apparatus may further comprise an information extractor/estimator for extracting or estimating the multiple signals needed for the second digital signal processing from the coded data after signal processing generated by the image coded data analyzer, and the image coded data synthesizer may generate the second image coded data by using the multiple signals output from the information extractor/estimator.
- This will offer an advantage of providing an image coded data re-encoding apparatus with a simple configuration because it obviates special processing involved in decoding.
- In the image coded data re-encoding apparatus, the image coded data synthesizer may generate the second image coded data with a data amount different from a data amount of the first image coded data input to the image coded data analyzer.
- This will offer an advantage of providing an image coded data re-encoding apparatus which can perform transformation between data of different amounts with reduced image degradation and processing delay.
- The image coded data re-encoding apparatus may further comprise a coefficient deletion/addition/correction portion for deleting part of the transform coefficients or quantization indices, which are extracted by the image coded data analyzer, and for correcting the transform coefficients or quantization indices in accordance with a ratio of amounts of data to be transformed.
- This will offer an advantage that the amount of data can be reduced because part of the transform coefficients or the quantization indices is deleted, and that the image quality of the decoded image can be improved as compared with a system which simply thins out the transform coefficients or quantization indices because the remainder of the transform coefficients or the quantization indices is corrected in accordance with the ratio of amounts of data to be transformed.
- The image coded data re-encoding apparatus may further comprise a coefficient deletion/addition/correction portion for deleting part of the transform coefficients or quantization indices, which are extracted by the image coded data analyzer, and are weighted in accordance with relationships between the transform coefficients or quantization indices and their neighboring transform coefficients or quantization indices.
- This will offer an advantage that the amount of data can be reduced, and the image quality of the decoded image can be improved as compared with a system which simply thins out the transform coefficients or quantization indices.
- The image coded data re-encoding apparatus may further comprise a coefficient deletion/addition/correction portion for adding, to the transform coefficients or quantization indices which are extracted in the image coded data analyzer, new transform coefficients or quantization indices after correcting the new transform coefficients or quantization indices in accordance with a ratio of amounts of data to be transformed.
- This will offer an advantage of achieving transform with improved image quality as compared with a system which simply adds the transform coefficients or quantization indices because the newly added transform coefficients or quantization indices are corrected in accordance with the ratio of amounts of data to be transformed, and hence the data amount of the additional data can be adjusted considering the image quality after inverse transform.
- The image coded data re-encoding apparatus may further comprise a coefficient deletion/addition/correction portion for adding, to the transform coefficients or quantization indices which are extracted in the image coded data analyzer, new transform coefficients or quantization indices after predicting transform coefficients or quantization indices including the new transform coefficients or quantization indices and their neighboring transform coefficients or quantization indices.
- This will offer an advantage of achieving transform with improved image quality, providing clearer images than a system which simply adds the transform coefficients or quantization indices, because such prediction is carried out as improving the decoded image quality in adding the transform coefficients or quantization indices.
- The image coded data re-encoding apparatus may further comprise a coefficient deletion/addition/correction portion for increasing a ratio of deletion of the transform coefficients or quantization indices, which are extracted by the image coded data analyzer, if the coding parameter designating the picture type indicates, when decision is made whether or not the current image to be processed is used for prediction in future coding, that the picture type is not used for the prediction in the future coding.
- This will offer an advantage of achieving high quality transform which can maintain the total image quality on the time axis because when a frame of unit time length is not used for the coding in the next unit time, only the data amount associated with the frame can be reduced.
- The image coded data re-encoding apparatus may further comprise a coefficient deletion/addition/correction portion for decreasing a ratio of deletion of the transform coefficients or quantization indices, which are extracted by the image coded data analyzer, if the coding parameter designating the picture type indicates, when decision is made whether or not the current image to be processed is used for prediction in future coding, that the picture type is used for the prediction in the future coding.
- This will offer an advantage of achieving high quality transform which can maintain the total image quality on the time axis because when a frame of unit time length is used for the coding in the next unit time, the reduction ratio of the data in the frame is decreased.
- The image coded data re-encoding apparatus may further comprise a coefficient deletion/addition/correction portion for increasing a ratio of deletion of the transform coefficients or quantization indices, which are extracted by the image coded data analyzer, if the coding parameter designating the predictive type of the image block indicates, when decision is made whether a current image to be processed is used for prediction in future coding by using the coding parameters generated by the image coded data analyzer, that the image block is not used for the prediction in the future coding, even if the coding parameter designating the picture type indicates that the picture type is used for the prediction in the future coding.
- This will offer an advantage of achieving transform with improved image quality on the time axis because the decision is made of an increase of the deletion ratio of the transform coefficients or the quantization indices by using the coded block information in addition to the coded picture information, and hence finer control becomes possible.
- In the image coded data re-encoding apparatus, the image coded data synthesizer may generate the second image coded data whose decoding procedure differs from a decoding procedure of the first image coded data input to the image coded data analyzer.
- This will offer an advantage of providing higher quality image after the transform when generating image coded data between different coding systems.
- The image coded data re-encoding apparatus may further comprise a coding parameter corrector/transformer for correcting and transforming the expression form of various types of coding parameters, which are extracted by the image coded data analyzer, from the expression form in the decoding procedure of the first image coded data to an expression form in the decoding procedure of the second image coded data.
- This will offer an advantage of providing a small size, inexpensive apparatus because it becomes unnecessary for the first image coded data to be once decoded into an image and then re-encoded in accordance with the coding system, even when the first image coded data input to the image coded data analyzer has different coding system from the second coded data output from the image coded data synthesizer.
- In the image coded data re-encoding apparatus, the image coded data synthesizer may generate the second image coded data including an image signal whose image size differs in time or space from an image size of an image signal included in the first image coded data input to the image coded data analyzer.
- This will offer an advantage of providing an image coded data re-encoding apparatus which facilitates the transform between the different image sizes in time or space, thereby achieving high quality image after the transform.
- The image coded data re-encoding apparatus, may further comprise a coefficient deletion/addition/correction portion for changing an amount of the transform coefficients or quantization indices extracted by the image coded data analyzer, and for correcting the transform coefficients or quantization indices, which are extracted by the image coded data analyzer, in accordance with a ratio of the image sizes to be transformed.
- This will offer an advantage of reducing the degradation of the image quality after the change in the image size by suppressing sharp degradation in the resolution or unnaturalness of the image, because the transform coefficients or the quantization indices are corrected in accordance with the image sizes to be varied in the coded data transform involving the image size change.
- The image coded data re-encoding apparatus may further comprise a coefficient deletion/addition/correction portion for correcting dimension of the motion vectors extracted by the image coded data analyzer in accordance with a ratio of the image sizes to be transformed.
- This will offer an advantage that a characteristic is obtained, which substantially matches the characteristic obtained in a wider motion compensative search range, by a narrower range motion compensative search, because of the improved motion compensation efficiency in the image coded data after the transform since the dimension of the motion vectors are corrected in accordance with the ratio of the image sizes to be changed.
- In the image coded data re-encoding apparatus, the image coded data synthesizer may generates the second image coded data including an image signal whose sequence differs from a sequence of an image signal included in the first image coded data input to the image coded data analyzer.
- This will offer an advantage of providing an image coded data re-encoding apparatus that can achieve transform between image coded data whose image signal sequences are different, thereby achieving high image quality after the transform.
- The image coded data re-encoding apparatus may further comprise a motion searcher for estimating dimension of the motion vectors extracted by the image coded data analyzer in accordance with the sequence of the image signals to be transformed.
- This will offer an advantage of improving the efficiency of coding using the motion vectors after the transform by estimating the dimension of the motion vectors in accordance with the sequences of the images to be transformed.
- In the image coded data re-encoding apparatus, the image coded data synthesizer may generate the second image coded data whose decoded image signal includes a number of frames per unit time different from a number of frames per unit time of a decoded image signal of the first image coded data input to the image coded data analyzer.
- This will offer an advantage of providing an image coded data re-encoding apparatus capable of implementing the transform into the image coded data whose number of image frames differs from that of the decoded image signal, thereby achieving high image quality after the transform.
- The image coded data re-encoding apparatus may further comprise a quantization estimator for estimating, from the decoded image output from the image coded data analyzer, quantization parameters obtained in the course of generating the first image coded data, and the image coded data synthesizer may generate the second image coded data by using the quantization parameters estimated by the quantization estimator.
- This is effective when the image coded data analyzer carries out operation matching a common decoding operation, and will offer an advantage of providing an image coded data re-encoding apparatus capable of maintaining the image quality after the transform in such application as relay transmission because the estimation of the quantization indices from the decoded image makes it possible to improve the image quality of the second image coded data. In addition, it offers an advantage of implementing an optimum transform by controlling the quantization parameters in accordance with the ratio of data rates before and after the transform. In particular, the best image quality can be achieved when the rates are identical before and after the transform.
- FIG. 1 is a block diagram showing a first embodiment of an image coded data re-encoding apparatus in accordance with the present invention;
- FIG. 2 is a block diagram showing a coding processor in the first embodiment;
- FIG. 3 is a schematic diagram illustrating a coding mode in the coding processor;
- FIG. 4 is a block diagram showing a decoding processor in the first embodiment;
- FIG. 5 is a block diagram showing a second embodiment of an image coded data re-encoding apparatus in accordance with the present invention;
- FIG. 6 is a block diagram showing a third embodiment of an image coded data re-encoding apparatus in accordance with the present invention;
- FIG. 7 is a block diagram showing fourth to seventh embodiments of an image coded data re-encoding apparatus in accordance with the present invention;
- FIG. 8 is a block diagram showing eighth and ninth embodiments of an image coded data re-encoding apparatus in accordance with the present invention;
- FIG. 9 is a block diagram showing a tenth embodiment of an image coded data re-encoding apparatus in accordance with the present invention;
- FIG. 10 is a block diagram showing an eleventh embodiment of an image coded data re-encoding apparatus in accordance with the present invention;
- FIG. 11 is a block diagram showing a twelfth embodiment of an image coded data re-encoding apparatus in accordance with the present invention;
- FIG. 12 is a block diagram showing a thirteenth embodiment of an image coded data re-encoding apparatus in accordance with the present invention;
- FIG. 13 is a block diagram showing a fourteenth embodiment of an image coded data re-encoding apparatus in accordance with the present invention;
- FIG. 14 is a block diagram showing a fifteenth embodiment of an image coded data re-encoding apparatus in accordance with the present invention;
- FIG. 15 is a block diagram showing a sixteenth embodiment of an image coded data re-encoding apparatus in accordance with the present invention; and
- FIG. 16 is a block diagram showing a conventional image coded data re-encoding apparatus.
- The invention will now be described with reference to the accompanying drawings.
-
Embodiment 1 - FIG. 1 is a block diagram showing a first embodiment of an image coded data re-encoding apparatus in accordance with the present invention, together with a coding processor and a decoding processor, wherein the coding processor codes an input image signal to be subjected to the digital signal processing in the image coded data re-encoding apparatus, and the decoding processor decodes the image coded data processed in the image coded data re-encoding apparatus. In this figure, the
reference numeral 30 designates an image coded data re-encoding apparatus, 40 designates a coding processor, and 50 designates a decoding processor. Thereference numeral 200 designates a digital input image signal supplied to thecoding processor input image signal 200 by thecoding processor data 220 in the image codeddata re-encoding apparatus data 240 in thedecoding processor 50. - In the image coded
data re-encoding apparatus 30, thereference numeral 310 designates an image coded data analyzer, and 320 designates an image coded data synthesizer. The image codeddata analyzer 310 performs a first digital signal processing on the first image codeddata 220 fed from thecoding processor 40. The image codeddata synthesizer 320 receives the coded data after the signal processing output from the image codeddata analyzer 310 together with multiple signals associated with the first image coded data, and performs a second digital signal processing on the coded data after the signal processing on the basis of the multiple signals, thereby generating a second image codeddata 240. Thereference numeral 221 designates the coded data after signal processing, which is transferred from the image coded data analyzer 310 to the image codeddata synthesizer data 220, which are also transferred from the image coded data analyzer 310 to the image codeddata synthesizer 320. Thereference numeral 223 designates information for instructing a data amount and an image size to be transformed, which is input to the image codeddata synthesizer 320. - In the
coding processor 40, thereference numeral 401 designates a transformer, 402 designates a quantizer, and 403 designates a variable length coder. Thetransformer 401 carries out operations such as discrete cosine transform (called DCT below) to generate transform coefficients from theinput image signal 200. Thequantizer 402 generates quantization indices by applying a scalar quantization processing to the transform coefficients generated by thetransformer 401. Thevariable length coder 403 performs variable length coding on the quantization indices generated by thequantizer 402 by using Huffman codes or the like. - In the
decoding processor 50, thereference numeral 501 designates a variable length decoder, 502 designates an inverse quantizer, and 503 designates an inverse transformer. Thevariable length decoder 501 applies the variable length decoding to the second image codeddata 240. Theinverse quantizer 502 performs inverse scalar quantization processing on the quantization indices decoded by thevariable length decoder 501. Theinverse transformer 503 performs the inverse DCT on the transform coefficients generated by the inverse quantization by theinverse quantizer 502 to obtain the decodedimage signal 250. In thecoding processor 40 anddecoding processor 50, only blocks are shown for implementing basic functions. - Next, the operation will be described.
- The digital
input image signal 200 is input to thecoding processor 40 comprising thetransformer 401,quantizer 402 andvariable length coder 403 to be coded into the first image codeddata 220. The operation of thecoding processor 40 will be described below. - FIG. 2 is a block diagram showing a concrete configuration of the
coding processor 40 to describe its operation. It includes thetransformer 401,quantizer 402 andvariable length coder 403 for implementing the basic function for coding theinput image signal 200 into the first image codeddata 220, together with an additional portion for implementing a motion compensative prediction. In this figure, thereference numeral 404 designates a current coded data frame memory for storing theinput image signal frame image 210 from acurrent frame image 201 read out of the current codeddata frame memory 404, thereby generating a predictionerror frame image 202. The predictedframe image 210 is fed from a motion compensative predictor which will be described later. Thereference numeral 203 designates the transform coefficients output from thetransformer 401 as a result of the DCT of the predictionerror frame image 202 by thetransformer 401. Thereference numerals transform coefficients 203 by thequantizer 402 and supplied to thevariable length coder 403. Thereference numeral 406 designates an inverse quantizer for carrying out the inverse quantization of thequantization indices 204 output from thequantizer transform coefficients 206 which is obtained by the inverse quantization by theinverse quantizer error frame image 207, which is generated by the inverse transform by theinverse transformer 407, to the predictedframe image 210 from the motion compensative predictor. Thereference numeral 409 designates a preceding coded data frame memory for storing a local decodedframe image 208 obtained as a result of addition by theadder frame image 210 delivered to thesubtracter 405 and theadder 408, andmotion vectors 211 supplied to thevariable length coder 403, on the basis of apreceding frame image 209 read out of the preceding codeddata frame memory 409 and thecurrent frame image 201 read out of the current codeddata frame memory 404. Thevariable length coder 403 generates the first image codeddata 220 using themotion vectors 211, thequantization parameters 205 from thequantizer 402, and thequantization indices 204 from thequantizer 402. - The
coding processor 40 with the configuration as shown in FIG. 2 operates as follows in accordance with the MPEG 1 (Moving Picture Expert Group 1) standard proposed by the joint conference of ISO (International Organization for Standardization) and IEC (International Electrotechnical Commission). First, the digitalinput image signal 200 is stored in the current codeddata frame memory 404, and thecurrent frame image 201 to be coded which is read out of the current codeddata frame memory 404 is input to thesubtracter 405. Thesubtracter 405 subtracts from the input signal the predictedframe image 210 fed from themotion compensative predictor 410, thereby generating the predictionerror frame image 202. Subsequently, thetransformer 401 transforms the predictionerror frame image 202 using the DCT to generate thetransform coefficients 203. Thequantizer 402 performs the scalar quantization processing in accordance with quantization steps using feedback control which monitors the amount of generated codes with reference to thetransform coefficients 203 to keep that amount constant, or using feed forward control based on the detected variance of theinput image signal 200, thereby generating thequantization indices 204. Thevariable length coder 403 performs the variable length coding on thequantization indices 204 using the Huffman codes together with thequantization parameters 205 generated by thequantizer 402 and themotion vectors 211 generated by themotion compensative predictor 410, thus generating the first image codeddata 220. The first image codeddata 220 is transmitted through a communication channel, or stored in a storage medium such as CD ROM (compact disk memory) or video tape. - The
inverse quantizer 406 performs inverse quantization of thequantization indices 204 generated by thequantizer 402, and outputs thetransform coefficients 206. Theinverse transformer 407 carries out on thetransform coefficients 206 the inverse transform using the inverse DCT to generate the local predictionerror frame image 207 fed to theadder 408. Theadder 408 adds the local predictionerror frame image 207 to the predictedframe image 210 fed from themotion compensative predictor 410 to generate the local decodedframe image 208 which is stored in the preceding codeddata frame memory 409. Subsequently, themotion compensative predictor 410 performs pattern matching of the coded precedingframe image 209 read out of the preceding codeddata frame memory 409 with thecurrent frame image 201 read out of the current codeddata frame memory 404 to generate the motion compensated predictedframe image 210 which provides minimum error, and supplies it to thesubtracter 405 and theadder 408. - In the
MPEG 1 motion compensative prediction, there are three types of coding modes as shown in FIG. 3: an intra-frame coding mode (I-Picture); a forward motion compensative inter-frame predictive coding mode (P-Picture); and a bidirectional motion compensative inter-frame predictive coding mode (B-Picture). With regard to I pictures (intra-frame coding mode frame images) F(0) and F(9), the motion compensative prediction is not carried out, that is, the predictedframe image 210 is not generated. Since I pictures are used as a reference image for the motion compensative prediction, a high quality decoded image is required of the I pictures. The amount of codes, on the other hand, considerably increases because the motion compensative prediction is not performed. With regard to P pictures (forward motion compensative inter-frame predictive coding mode frame images) F(3) and F(6), the motion compensative prediction is carried out using only a preceding image, such as F(0) for F(3), F(3) for F(6). Since the P pictures are occasionally used as the reference image of the motion compensative prediction, their quality must be kept at a certain degree. With regard to B pictures (bidirectional motion compensative inter-frame predictive coding mode frame images) F(1) and F(2), F(4) and F(5), and F(7) and F(8), the motion compensative prediction is performed using two images, preceding and following ones. Since the B pictures are not used as the reference image of the motion compensative prediction, rough quantization is possible. The bidirectional motion compensative prediction can reduce the code amount because the decoded images can be obtained only from the motion vectors if the preceding and following I pictures and P pictures have high quality in a stream of motion sequences. - When the current coded
data frame memory 404 takes the coding mode as shown in FIG. 3, it outputs the frame images in the order of F(0)-F(3)-F(1)-F(2)-F(6)-F(4)-F(5)-F(9)-F(7)-F(8). - The image coded
data re-encoding apparatus 30 generates the second image codeddata 240 from the first image codeddata 220 thecoding processor 40 has generated by applying the coding on theinput image signal 200 by transforming at least one of the “volume of the coded data”, “size of the coded image”, “coding systems” and “sequence of the images”. Specifically, the image codeddata analyzer 310 receives the first image codeddata 220 from thecoding processor 40, and analyzes the data to generate the coded data aftersignal processing 221 after the coding. In addition, it extracts themultiple signals 222 associated with the first image coded data in the process of generating the coded data aftersignal processing 221, thereby outputting the multiple signals. Subsequently, the image codeddata synthesizer 320 generates the second image codeddata 240 from the coded data aftersignal processing 221 and themultiple signals 222 associated with the first image coded data, and outputs it to thedecoding processor 50. Thus, the image codeddata synthesizer 320 is paired with the image codeddata analyzer 310. Considering the image coded data analyzer 310 as an integral part of an image decoder, which carries out the decoding of the first image codeddata 220 halfway, the image codeddata synthesizer 320 can be considered as an integral part of an image coder, which performs the re-encoding of the coded data aftersignal processing 221 which has been decoded halfway by the image codeddata analyzer 310. - This will be described in more detail with reference to the drawings.
- The image coded data analyzer310 of the image coded
data re-encoding apparatus 30 receives the first image codeddata 220 which is obtained by coding in thecoding processor 40 the digitalinput image signal 200 like a digital motion image signal. The image codeddata analyzer 310 has a function to perform the variable length decoding of the first image codeddata 220, and a function to carry out the inverse quantization of the decoded quantization indices, and analyzes the first image codeddata 220 when it is input. Specifically, it carries out a first digital signal processing on the input first image codeddata 220 by using the variable length decoding and inverse quantization functions, thereby supplying the image codeddata synthesizer 320 with results of the first digital signal processing, that is, the coded data aftersignal processing 221 and themultiple signals 222 associated with the first image coded data, which has been extracted from the first image codeddata 220 during the first digital signal processing. - On the other hand, the image coded
data synthesizer 320 has a function to perform on the coded data aftersignal processing 221 fed from the image coded data analyzer 310 a coefficient correction that deletes, adds, or corrects the transform coefficients of the coded data aftersignal processing 221. The image codeddata synthesizer 320 has another function to quantize the data which has undergone the coefficient correction, and still another function to perform the variable length coding of the quantization indices. The image codeddata synthesizer 320, receiving the coded data aftersignal processing 221, themultiple signals 222 associated with the first image coded data, and theinformation 223 instructing the transform, carries out the synthesis of the coded data by using the functions such as the coefficient correction, quantization and variable length coding. Specifically, it performs a second digital signal processing of the coded data aftersignal processing 221 based on themultiple signals 222 associated with the first image coded data and theinformation 223 instructing the transform, thereby generating the second image codeddata 240 as a result of the second digital signal processing. In this process, themultiple signals 222 associated with the first image coded data, which is extracted from the first image codeddata 220 during the first digital signal processing by the image codeddata analyzer 310, is used when the first image codeddata 220 is transformed, without being decoded to an image, to the second image codeddata 240. The second image codeddata 240 thus generated is transmitted through a communication channel, or recorded on a storage medium such as a CD-ROM, or a video tape, to be supplied to thedecoding processor 50. - The second image coded
data 240 as transmitted/recorded information, which has been transmitted through the communication channel or recorded on the storage medium such as CD-ROM or video tape, is decoded by thedecoding processor 50 to be output as the decodedimage signal 250. In other words, thedecoding processor 50 provides an inverse process of thecoding processor 40. - FIG. 4 is a block diagram illustrating a concrete configuration of the
decoding processor 50 to explain its operation. In this figure, thevariable length decoder 501,inverse quantizer 502 andinverse transformer 503 function as an inverse transformer which implements the basic function for decoding the second image codeddata 240 into the decodedimage signal 250. The remaining portion is added to implement motion compensative prediction. In FIG. 4, thereference numerals data 240 by thevariable length decoder 501. Thereference numeral 244 designates transform coefficients generated by the inverse quantization of thequantization indices 241 by theinverse quantizer transform coefficients 244 by theinverse transformer 503. Thereference numeral 504 designates an adder for adding the predictionerror frame image 245 to a predictedframe image 248 fed from the motion compensative predictor described later, and 246 designates a current frame image output from theadder 504. Thereference numeral 505 designates a current decoded data frame memory, into which thecurrent frame image 246 is stored, and from which thecurrent frame image 246 is read as a decodedimage signal 250. Thereference numeral 506 designates a preceding decoded data frame memory, into which thecurrent frame image 246 is stored, and from which it is read in the next cycle as a precedingframe image 247. Thereference numeral 507 designates a motion compensative predictor for generating a predictedframe image 248, which is supplied to theadder 504, from the precedingframe image 247 and thecurrent frame image 246 on the basis of the decodedmotion vectors 243 from thevariable length decoder 501. - The
variable length decoder 501 of thedecoding processor 50 with the arrangement as shown in FIG. 4 receives the second image codeddata 240 which has been generated by applying the first and second digital signal processings to the first image codeddata 220 in the image codeddata re-encoding apparatus 30, and recorded on the storage medium such as the CD-ROM or the video tape. Thevariable length decoder 501 performs the variable length decoding of the input second image codeddata 240 to decode it to thequantization indices 241 and to generate thequantization parameters 242 and decodedmotion vectors 243. Subsequently, theinverse quantizer 502 performs the inverse scalar quantization of thequantization indices 241 decoded by thevariable length decoder 501 in accordance with thequantization parameters 242 from thevariable length decoder 501, thereby obtaining the inverse quantizedtransform coefficients 244. Then, theinverse transformer 503 carries out the inverse DCT of thetransform coefficients 244 which has undergone the inverse quantization in theinverse quantizer 502, thus to generate the predictionerror frame image 245. Theadder 504 adds the predictionerror frame image 245 which has undergone the inverse transform in theinverse transformer 503 to the predictedframe image 248 fed from themotion compensative predictor 507 to generate thecurrent frame image 246 to be decoded. Thecurrent frame image 246 is stored in the current decodeddata frame memory 505, and is read therefrom as the decodedimage signal 250. Thecurrent frame image 246 is also stored in the preceding decodeddata frame memory 506 so that it is read during the motion compensative prediction in the next cycle as the precedingframe image 247 which has already been decoded, and is supplied to themotion compensative predictor 507. Themotion compensative predictor 507 generates in accordance with the decodedmotion vectors 243 fed from thevariable length decoder 501 the predictedframe image 248 to be supplied to theadder 504 from the precedingframe image 247 which has already been decoded and thecurrent frame image 246 which is output from theadder 504. - Although in the above description, the motion compensative prediction is carried out in the
coding processor 40 and thedecoding processor 50, the motion compensative prediction may be omitted. - According to the first embodiment, the
multiple signals 222 associated with the first image coded data is employed in the second digital signal processing for generating the second image codeddata 240, whichmultiple signals 222 are extracted in the process of applying the first digital signal processing to the first image codeddata 220. This makes it unnecessary to add special information for the second digital signal processing, and hence reduces the information amount, resulting in information efficient apparatus. -
Embodiment 2 - Although the foregoing
embodiment 1 used themultiple signals 222 associated with the first image coded data, which are extracted in the process of applying the first digital signal processing to the first image codeddata 220, to generate the secondcoded data 240 by applying the second digital signal processing to the coded data aftersignal processing 221, the multiple signals associated with the first image coded data can be obtained by other ways. For example, information which is used for generating the first image codeddata 220 by coding theinput image signal 200 in thecoding processor 40, and which cannot be extracted in the process of applying the first digital signal processing to the first image codeddata 220 in the image codeddata re-encoding apparatus 30, can be combined with the first image codeddata 220 to be fed to the image codeddata re-encoding apparatus 30 so that the image codeddata re-encoding apparatus 30 separates the information from the first image codeddata 220 as the multiple signals associated with the first image coded data which are used for performing the second digital signal processing on the first image codeddata 220. - FIG. 5 is a block diagram showing an
embodiment 2 of the image codeddata re-encoding apparatus 30 in accordance with the present invention together with thecoding processor 40 and thedecoding processor 50, wherein the corresponding portions are designated by the same reference numerals as in FIG. 1, and the description thereof is omitted here. In FIG. 5, thereference numeral 414 designates a combiner provided in thecoding processor 40 for combining, with the first image codeddata 220 which is output from thevariable length coder 403, the above-mentioned information which is used by thetransformer 401 and thequantizer 402 for generating the first image codeddata 220 by coding theinput image signal 200 in thecoding processor 40, and which cannot be extracted in the process of applying the first digital signal processing to the first image codeddata 220 in the image codeddata re-encoding apparatus combiner 414. Thereference numeral 340 designates a separator provided in the image codeddata re-encoding apparatus 30 for separating from thecombination data 260 delivered from thecoding processor 40 the first image codeddata 220 and the information that cannot be extracted in the first digital data processing, and 224 designates multiple signals associated with the first image coded data corresponding to the information which cannot be extracted in the first digital signal processing. The first image codeddata 220 and themultiple signals 224 associated with the first image coded data, which are separated by theseparator 340, are supplied to the image codeddata analyzer 310 and the image codeddata synthesizer 320, respectively. - Next, the operation of the
embodiment 2 will be described. - The
coding processor 40 basically operates in the same manner as that of the embodiment 1: It performs the coding of the digitalinput image signal 200 to obtain the first image codeddata 220. Only, it differs from theembodiment 1 in the following: The information which is used for the coding by thetransformer 401 and thequantizer 402, and which cannot be extracted in the process of performing the first digital signal processing on the first image codeddata 220 by the image codeddata re-encoding apparatus 30, is sent to thecombiner 414. Thecombiner 414 combines the information with the first image codeddata 220 output from thevariable length coder 403 to generate thecombination data 260, and supplies it to the image codeddata re-encoding apparatus 30. Here, the information which is output from thetransformer 401 and thequantizer 402, and which cannot be extracted in the process of applying the first digital signal processing to the first image codeddata 220, differs depending on the coding method employed. For example, whenMPEG 1 is employed, the parameter used for specifying the type of thetransformer 401 is output from thetransformer 401, and the parameters used for specifying the characteristic of thequantizer 402 are output from thequantizer 402. - In the image coded
data re-encoding apparatus 30, receiving thecombination data 260 from thecoding processor 40, theseparator 340 separates it to the first image codeddata 220 and themultiple signals 224 concerning the first image coded data associated with the information which cannot be extracted in the process of the first digital signal processing. The divided first image codeddata 220 is supplied from theseparator 340 to the image codeddata analyzer 310. The image coded data analyzer 310 carries out the first digital signal processing of the first image codeddata 220, and supplies the image codeddata synthesizer 320 with the result of the first digital signal processing as the coded data aftersignal processing 221. - Accordingly, in this case also, the image coded
data analyzer 310 can be considered as a part of the image decoder. On the other hand, themultiple signals 224 concerning the first image coded data, which are separated by theseparator 340, are input to the image codeddata synthesizer 320. The image codeddata synthesizer 320 paired with the image codeddata analyzer 310, receiving the coded data aftersignal processing 221 and themultiple signals 224 concerning the first image coded data, carries out the second digital signal processing in accordance with theinformation 223 that commands the transform and defines the data amount and image size to be transformed. Accordingly, in this case also, the image codeddata synthesizer 320 can be considered as a part of the image coder. The image synthesis processing of the coded data in the image codeddata synthesizer 320 is performed in a manner similar to that of theembodiment 1 except that themultiple signals 224 concerning the first image coded data, which are separated from thecombination data 260 by theseparator 340, are used instead of themultiple signals 222 associated with the first image coded data, which are extracted by the image codeddata analyzer 310. Themultiple signals 224 concerning the first image coded data, which are used for transforming the first image codeddata 220 into the second image codeddata 240 without once decoding the first image codeddata 220 into an image, also serves as the information for achieving improved images as compared with the case where themultiple signals 222 associated with the first image coded data is used. The second image codeddata 240 generated as the result of the second digital signal processing is transmitted through a communication channel or recorded on the storage medium such as a CD-ROM or video tape, and is input to thedecoding processor 50 which decodes it to the decodedimage signal 250 in exactly the same way as thedecoding processor 50 of theembodiment 1. - Thus, the
embodiment 2 uses, for the second digital signal processing in the image codeddata synthesizer 320, themultiple signals 224 which cannot be extracted in the process of the first digital signal processing. As a result, it has an advantage of achieving more efficient transform than an apparatus which does not use themultiple signals 224. -
Embodiment 3 - The foregoing embodiments generate the second image coded
data 240 by applying the second digital signal processing to the coded data aftersignal processing 221 by using themultiple signals data 220, multiple signals associated with the first image codeddata 220, which are extracted or estimated from the coded data aftersignal processing 221 as information needed for re-encoding the coded data aftersignal processing 221. In this case, the coded data aftersignal processing 221 is obtained by decoding the first image codeddata 220 through the first digital signal processing using the image coded data analyzer 310 as an image decoder and the image codeddata synthesizer 320 as an image coder. - FIG. 6 is a block diagram showing a third embodiment of the image coded
data re-encoding apparatus 30 in accordance with the present invention, together with thecoding processor 40 anddecoding processor 50, in which the corresponding portions to those of theembodiment 1 are designated by the same reference numerals, and the description thereof is omitted here. In this figure, thereference numeral 350 designates an information extractor/estimator, and 225 designates multiple signals associated with the first image coded data output from the information extractor/estimator 350. The information extractor/estimator 350 extracts or estimates from the coded data aftersignal processing 221 the information which is needed for re-encoding the coded data aftersignal processing 221 in the second digital signal processing, and supplies the resultant information to the image codeddata synthesizer 320 as themultiple signals 225 associated with the first image coded data. In this case, the coded data aftersignal processing 221 is obtained by decoding the first image codeddata 220 by applying the first digital signal processing to the image codeddata 220 in the image codeddata analyzer 310. - Next, the operation will be described.
- The description is omitted here of the process of generating the first image coded
data 220 by coding theinput image signal 200 with thecoding processor 40, and the process of generating the decodedimage signal 250 by decoding the second image codeddata 240 with thedecoding processor 50, because they are the same as those of theembodiment 1. Thus, the operation of only the image codeddata re-encoding apparatus 30 will be described here. - In the image coded
data re-encoding apparatus 30, receiving the first image codeddata 220 from thecoding processor 40, the image codeddata analyzer 310 performs the first digital signal processing of the first image codeddata 220 to decode it. The decoded image data is fed to the information extractor/estimator 350 and the image codeddata synthesizer 320 as the coded data aftersignal processing 221. The information extractor/estimator 350 extracts or estimates from the coded data aftersignal processing 221 fed from the image coded data analyzer 310 themultiple signals 225 associated with the first image coded data, which are needed for the second digital signal processing in the image codeddata synthesizer 320, and supplies them to the image codeddata synthesizer 320. The image codeddata synthesizer 320 is provided with theinformation 223 which defines the data amount and image size to be transformed and commands the transform, in addition to themultiple signals 225 associated with the first image coded data, and the coded data aftersignal processing 221 from the image codeddata analyzer 310. The image codeddata synthesizer 320 applies the second digital signal processing to the coded data aftersignal processing 221 to generate the second image codeddata 240. In this case, themultiple signals 225 associated with the first image coded data which are extracted or estimated from the coded data aftersignal processing 221 by the information extractor/estimator 350 are not only used for decoding the first image codeddata 220 into an image, but also used for generating the second image codeddata 240. The thus generated second image codeddata 240 is fed from the image codeddata re-encoding apparatus 30 to thedecoding processor 50. - The
embodiment 3 with such an arrangement has an advantage that the configuration of the image codeddata re-encoding apparatus 30 is simplified. This is because no special processing is required for decoding since themultiple signals 225 associated with the first image codeddata 220, which are needed for the second digital signal processing by the image codeddata synthesizer 320, are extracted or estimated from the coded data aftersignal processing 221 which are once decoded from the first image codeddata 220. -
Embodiment 4 - FIG. 7 is a block diagram showing the internal configuration of the image coded
data analyzer 310 and the image codeddata synthesizer 320 as anembodiment 4 of the image codeddata re-encoding apparatus 30 in accordance with the present invention, which can reduce the amount of data between the first image codeddata 220 and the second image codeddata 240. In this figure, thereference numeral 311 designates a variable length decoder for carrying out variable length decoding of the first image codeddata 220 delivered from thecoding processor variable length decoder 311. Thereference numeral 312 designates an inverse quantizer for performing inverse quantization of thequantization indices inverse quantizer 312. The inversely quantized transformcoefficients 227 is output as the coded data aftersignal processing 221. The image coded data analyzer 310 in accordance with thepresent embodiment 4 includes thesevariable length decoder 311 andinverse quantizer 312. - The
reference numeral 321 designates a coefficient deletion/addition/correction portion for deleting, adding or correcting a part of the inversely quantizedtransform coefficients 227 delivered from the image coded data analyzer 310 as the coded data aftersignal processing 221 in response to theinformation 223 that commands the transform and defines the data amount to be transformed. Thereference numeral 228 designated corrected transform coefficients, the result of the coefficient correction by the coefficient deletion/addition/correction portion quantizer 322. Thereference numeral 323 designates a variable length coder which codes thequantization indices 229, and supplies thedecoding processor 50 with the resultant second image codeddata 240. The image codeddata synthesizer 320 in accordance with theembodiment 4 includes the coefficient deletion/addition/correction portion 321,quantizer 322 andvariable length coder 323. - Next, the operation will be described.
- It is assumed here, that the
coding processor 40 and thedecoding processor 50 employ a coding system based on the transform coding including the coding and quantization as a basic function, and that the image codeddata analyzer 310 performs a part of the decoding, wherein thecoding processor 40 generates the first image codeddata 220 by coding theinput image signal 200, and supplies it to the image codeddata re-encoding apparatus 30, and thedecoding processor 50 generates the decodedimage signal 250 by decoding the second image codeddata 240 the image codeddata re-encoding apparatus 30 outputs. In this case, the image codeddata analyzer 310 extracts the transform coefficients or the quantization indices by carrying out the inverse quantization of the first image codeddata 220, and the image codeddata synthesizer 320 deletes a part of the inverse quantized transform coefficients or the quantization indices, and corrects the remaining transform coefficients and the quantization indices in accordance with the ratio of the amount of the data to be transformed. This can facilitate the transform considering the image quality after the inverse transform as compared with the case where part of the transform coefficients or the quantization indices is simply deleted. - Referring the drawings, the operation will be described in more detail.
- The
variable length decoder 311, receiving the first image codeddata 220, carries out the variable length decoding referring a table prepared in advance to produce thecombination data 260 corresponding t the first image codeddata 220. An example of the table is shown in Table 1.TABLE 1 occurrence probability data code (coded data) 0.3 1 00 0.25 2 01 0.2 3 11 0.1 4 101 0.08 5 1000 0.07 6 1001 - In this table, each of six data is assigned a code corresponding to the possibility of occurrence such that a shorter code is assigned to data with higher possibility of occurrence, and vice versa. Thus, assignment of codes with different length is performed in accordance with the possibility of occurrence of the quantization indices. The
variable length decoder 311 carries out the variable length decoding by extracting data corresponding to the code (coded data) from the table. - Thus, the
quantization indices 226 output from thevariable length decoder 311 are input to theinverse quantizer 312. Theinverse quantizer 312 generates the inversely quantizedtransform coefficients 227 by performing the inverse quantization of thequantization indices 226 in accordance with the quantization parameters (not shown), and supplies it to the image codeddata synthesizer 320 as the coded data aftersignal processing 221. Incidentally, the quantization in the digital processing can usually be achieved by division, whereas the inverse quantization is achieved by multiplication. - The coefficient deletion/addition/
correction portion 321 of the image codeddata synthesizer 320, receiving theinformation 223 which commands the transform, deletes a part of the coded data aftersignal processing 221 fed from the image coded data analyzer 310 in accordance with the data amount to be transformed instructed by theinformation 223 commanding transform, and carries out the coefficient correction of the remainder of the coded data in accordance with the ratio of the data amount to be transformed which is defined by theinformation 223 commanding transform. - In this
embodiment 4 of the image codeddata re-encoding apparatus 30, the data amount is reduced through the transform. When reducing the data amount of the transform coefficients, it is advantageous to start deletion of the transform coefficients from higher frequency components because lower frequency components are more significant than the higher frequency components of the transform coefficients. In this case, although simple deletion of the data in a particular region of the transform coefficients is effective for reducing the data amount itself, this will effect on the quality of images decoded from the transform coefficients with the reduced data amount. To prevent this, thepresent embodiment 4 not only deletes part of thetransform coefficients 227, but also corrects the remainder of thetransform coefficients 227 in accordance with the ratio of the data amount to be transformed which is instructed by theinformation 223 commanding transform. In this process, the effect on the decoded image the deleted part will produce is predicted, so that the correction of the remainder of thetransform coefficients 227 is achieved such that the effect becomes minimum. - The coefficient deletion/addition/
correction portion 321 deletes the part of thetransform coefficients 227 in this way, and the correctedtransform coefficients 228 generated by correcting the remainder are input to thequantizer 322. Thequantizer 322 carries out the quantization of the correctedtransform coefficients 228 to generate thequantization indices 229, and supplies them to thevariable length coder 323. Thevariable length coder 323 performs the variable length coding of thequantization indices 229 in accordance with the quantization parameters output from thequantizer 322 to generate the second image codeddata 240, and delivers it to thedecoding processor 50. - According to the
embodiment 4, the deletion of the part of thetransform coefficients 227, which are transferred from the image coded data analyzer 310 as the coded data aftersignal processing 221, enables the data amount to be reduced. In addition, the correction of the remainder of thetransform coefficients 227 in accordance with the ratio of its data amount enables the decoded image quality to be improved as compared with that obtained by simply thinning out the transform coefficients. -
Embodiment 5 - In the
embodiment 4, the data amount is reduced between the first image codeddata 220 and the second image codeddata 240 by deleting a part of the reversely quantized transform coefficients or the quantization indices, and by correcting the transform coefficients in accordance with the amount of data to be transformed. A part of the transform coefficients or the quantization indices, however, may be deleted by weighting the transform coefficients or the quantization indices to be reduced with the neighboring transform coefficients or the quantization indices, which is implemented in theembodiment 5. - In the
embodiment 5 of the image coded data re-encoding apparatus, the image coded data analyzer 310 as shown in FIG. 7 carries out the processings as follows: First, thevariable length decoder 311 decodes the first image codeddata 220 into thequantization indices 226. Then, theinverse quantizer 312 performs the inverse quantization of the decodedquantization indices 226, and transfers theresultant transform coefficients 227 to the image codeddata synthesizer 320 as the coded data aftersignal processing 221. The image codeddata synthesizer 320 receives thetransform coefficients 227 with the coefficient deletion/addition/correction portion 321 which deletes part of thetransform coefficients 227 after weighting thetransform coefficients 227 with its neighboring transform coefficients. Specifically, the coefficient deletion/addition/correction portion 321 examines, before deleting part of thetransform coefficients 227, the relationship between the transform coefficients to be deleted and their neighboring transform coefficients, performs weighting such that the deletion has a minimum effect on the quality of the decoded image, and deletes the transform coefficients to be deleted. - The
embodiment 5 also, as theembodiment 4, enables the decoded image quality to be improved as compared with that obtained by simply thinning out the transform coefficients. -
Embodiment 6 - Although the data amount is reduced between the first image coded
data 220 and the second image codeddata 240 in theembodiments embodiment 6. - In the
embodiment 6 of the image coded data re-encoding apparatus, the image coded data analyzer 310 as shown in FIG. 7 carries out the processings as follows: First, thevariable length decoder 311 decodes the first image codeddata 220 into thequantization indices 226. Then, theinverse quantizer 312 performs the inverse quantization of the decodedquantization indices 226, and transfers theresultant transform coefficients 227 to the image codeddata synthesizer 320. The image codeddata synthesizer 320 receives thetransform coefficients 227 with the coefficient deletion/addition/correction portion 321 which carries out the addition of new transform coefficients to increase the amount of data of the second image codeddata 240 by using thetransform coefficients 227 and a picture type which is obtained in the decoding procedure of the image coded data. Specifically, the coefficient deletion/addition/correction portion 321 carries out the correction of the newly added transform coefficients such that the effect of addition on the quality of the decoded image becomes minimum, in accordance with the ratio of the data amount defined by the information that commands the transform. - Considering the effect on the quality of the decoded image of increasing the data amount between the first image coded
data 220 and the second image codeddata 240, it will be advantageous to add the high frequency components of the transform coefficients or the quantization indices as in the case of reducing the data amount. Furthermore, the addition of the transform coefficients or the quantization indices in the higher frequency domain induces a greater change in the data amount than that in the lower frequency domain. - Thus, the
embodiment 6 corrects the newly added transform coefficients or the quantization indices in accordance with the ratio of the amount of data to be transformed. This makes it possible to add the data considering the image quality after the inverse transform, thereby achieving a higher image quality than that obtained by simply thinning out the transform coefficients. -
Embodiment 7 - The
embodiment 6 corrects the newly added transform coefficients or the quantization indices in accordance with the ratio of the amount of data to be transformed, when adding the transform coefficients or the quantization indices to the inversely quantized transform coefficients or the quantization indices to increase the amount of data between the first image codeddata 220 and the second image codeddata 240. The addition of the transform coefficients or quantization indices may be performed after predicting the transform coefficients or quantization indices including the newly added transform coefficients or quantization indices and their neighboring transform coefficients or quantization indices, which is implemented in theembodiment 7. - In the
embodiment 7 of the image coded data re-encoding apparatus, the image coded data analyzer 310 as shown in FIG. 7 carries out the processings as follows: First, thevariable length decoder 311 decodes the first image codeddata 220 into thequantization indices 226. Then, theinverse quantizer 312 performs the inverse quantization of the obtainedquantization indices 226, and transfers theresultant transform coefficients 227 to the image codeddata synthesizer 320. The image codeddata synthesizer 320 receives thetransform coefficients 227 with the coefficient deletion/addition/correction portion 321 which carries out prediction of the transform coefficients including the newly added transform coefficients and their neighboring transform coefficients, followed by the addition of the transform coefficients, when increasing the data amount of the second image codeddata 240 after the transform by using the receivedtransform coefficients 227. - Thus, the
embodiment 7 adds the transform coefficients or quantization indices after predicting the transform coefficients or quantization indices to be added and their neighbors. This makes it possible to implement more vivid images, providing an advantage of achieving the transform resulting in higher image quality than that obtained by the transform which simply thins out the transform coefficients. -
Embodiment 8 - FIG. 8 is a block diagram showing the internal configuration of the image coded
data analyzer 310 and the image codeddata synthesizer 320 of an eighth embodiment of the image codeddata re-encoding apparatus 30 in accordance with the present invention, which reduces the data amount between the first image codeddata 220 and the second image codeddata 240. In this figure, thereference numeral 230 designates coded picture information as a coding parameter indicating the picture type. Thecoded picture information 230 is output as the result of the decoding from thevariable length decoder 311 which performs the variable length decoding of the first image codeddata 220 supplied from thecoding processor 40. The remaining portions are the same as those of FIG. 7, and the description thereof is omitted here. Thevariable length decoder 311, however, differs from that of theembodiment 4 in that it outputs thecoded picture information 230 besides thequantization indices 226. In addition, the coefficient deletion/addition/correction portion 321 differs from that of theembodiment 4 in that it deletes, adds or corrects part of thetransform coefficients 227 by using the codedpicture information 230 besides the amount of data to be transformed which is instructed by theinformation 223 commanding the transform. Thecoded picture information 230 here is a parameter multiplexed picture by picture to represent the attributes such as the size of the picture, coding type of the picture (that is, the intra-frame coding, unidirectional motion compensative inter-frame predictive coding, or bidirectional motion compensative inter-frame predictive coding), maximum value of vectors used by the picture, and adaptive parameters used by the picture. - The operation will now be described.
- The eighth embodiment of the image coded data re-encoding apparatus implements the transform that can reduce the total degradation of the image quality by increasing the ratio of deletion of the transform coefficients or quantization indices included in the picture types which will not be used in the future prediction. This is performed by using the transform coefficients or quantization indices obtained by the inverse quantization of the first image coded
data 220 in the image codeddata analyzer 310, and the coded picture parameter obtained in the course of decoding the first image codeddata 220. - In the image coded data analyzer310 as shown in FIG. 8, the
variable length decoder 311 provides theinverse quantizer 312 with thequantization indices 226 obtained as a result of decoding the first image codeddata 220, and supplies the image codeddata synthesizer 320 with thecoded picture information 230 obtained in the course of the decoding. Theinverse quantizer 312 carries out the inverse quantization of the receivedquantization indices 226, and sends theresultant transform coefficients 227 to the image codeddata synthesizer 320 as the coded data aftersignal processing 221. In the image codeddata synthesizer 320, on the other hand, the coefficient deletion/addition/correction portion 321 receives theinformation 223, which commands the transform and defines the amount of data to be transformed, and deletes part of the receivedtransform coefficients 227. When theinformation 223 indicates the coded picture which will not be used for the prediction in the future coding, the ratio of deletion of the transform coefficients is increased. The received codedpicture information 230 includes, as described above, the coding type of the picture. When the coding type indicates the bidirectional motion compensative inter-frame prediction coding according to theMPEG 1, for example, the picture information is not used for the future coding. Thus, even if the reduction of the data amount results in the degradation of the image quality, there is no fear that it will continue. As a result, an increase in the deletion ratio of the transform coefficients in such picture information will improve the efficiency of the reduction of the data amount. - Thus, according to the
embodiment 8, an advantage is obtained that the data reduction can be implemented with high total efficiency by further reducing the data amount when the coded frame is not used in the next coding interval. -
Embodiment 9 - Although the
embodiment 8 handles the case in which the reduction ratio of the transform coefficients is increased when theinformation 223 indicates that the coded picture will not be used in the future prediction, the reduction ratio of the transform coefficients may be reduced when the information indicates that the coded picture will be used in the following prediction. More specifically, when the picture type indicates the use in the future coding, the reduction ratio of the transform coefficients or quantization indices will be reduced by using the transform coefficients or quantization indices obtained by the inverse quantization of the first image codeddata 220 in the image coded data analyzer 310 as shown in FIG. 8, and by using the picture type obtained in the course of decoding the first image codeddata 220. Thus, the transform is implemented which can prevent the degradation from being continued to the future by the prediction. - As a result, the
embodiment 9 has an advantage of achieving data reduction with higher total efficiency by decreasing the reduction in the amount of data of the coded data, when the coded frame is used in the next coding interval. -
Embodiment 10 - FIG. 9 is a block diagram showing the internal configuration of the image coded
data analyzer 310 and image codeddata synthesizer 320 of anembodiment 10 of the image codeddata re-encoding apparatus 30 in accordance with the present invention, which reduces the amount of data between the first image codeddata 220 and the second image codeddata 240. In this figure, thereference numeral 231 designates coded block information, one of the coding parameters indicating the predictive type of the image block. Thecoded block information 231 is output as the result of decoding by thevariable length decoder 311 which carries out the variable length decoding of the first image codeddata 220 supplied from thecoding processor 40. The remaining portions are designated by the same reference numerals as in FIG. 8, and the description thereof is omitted here. Thevariable length decoder 311, however, differs from that of theembodiment 8 in that it outputs thecoded block information 231 besides thequantization indices 226 and thecoded picture information 230. In addition, the coefficient deletion/addition/correction portion 321 differs from that of theembodiment 8 in that it deletes, adds or corrects part of thetransform coefficients 227 by using the codedblock information 231 besides thecoded picture information 230 and the amount of data to be transformed, which is instructed by theinformation 223 commanding the transform. Thecoded block information 231 is the information about the coded block (often called micro-block in the MPEG 1) which is the minimum unit of coding and represents quantization parameters, motion vectors, and the presence or absence of the motion compensative prediction, which are used in the coded block. - Next, the operation will be described.
- The tenth embodiment of the image coded data re-encoding apparatus implements the transform that can reduce the total degradation of the image quality by increasing the ratio of deletion of the transform coefficients or quantization indices included in the image blocks which are not used in the prediction, even if they belong to the picture types which will be used in the future coding. This is performed by using the transform coefficients or quantization indices obtained by the inverse quantization of the first image coded
data 220 in the image codeddata analyzer 310, and the picture types obtained in the course of decoding the first image codeddata 220. - In the image coded data analyzer310 as shown in FIG. 9, the
variable length decoder 311 provides theinverse quantizer 312 with thequantization indices 226 obtained as a result of decoding the first image codeddata 220, and supplies the image codeddata synthesizer 320 with thecoded picture information 230 andcoded block information 231 which are obtained in the course of the decoding. Theinverse quantizer 312 carries out the inverse quantization of the receivedquantization indices 226, and sends theresultant transform coefficients 227 to the image codeddata synthesizer 320 as the coded data aftersignal processing 221. In the image codeddata synthesizer 320, on the other hand, the coefficient deletion/addition/correction portion 321 receives thetransform coefficients 227,coded picture information 230 andcoded block information 231 besides theinformation 223, which commands the transform and defines the amount of data to be transformed, and deletes part of the receivedtransform coefficients 227. In the course of this, the coefficient deletion/addition/correction portion 321 decides on the basis of the receivedcoded block information 231 whether the information of the image block is used in the future coding, not on the coded picture basis but on the image block basis which enables finer control. If the result of this indicates that the image block is not used in prediction in the future coding, even though the coded picture will be used in the prediction in the future coding, the deletion ratio of the transform coefficients is increased. - Thus, according to the
embodiment 10, an advantage is obtained that the control based on a finer unit becomes possible and higher quality transform is achieved, because the decision is made whether the deletion ratio of the transform coefficients or quantization indices should be increased or not by using the codedblock information 231 in addition to thecoded picture information 230. - Embodiment 11
- FIG. 10 is a block diagram showing the internal configuration of the image coded
data analyzer 310 and image codeddata synthesizer 320 of an embodiment 11 of the image codeddata re-encoding apparatus 30 in accordance with the present invention, in which the decoding procedure of the first image codeddata 220 differs from that of the second image codeddata 240. In this figure, thereference numeral 311 designates a variable length decoder which carries out the variable length decoding of the first image codeddata 220 supplied from thecoding processor 40. Thereference numeral 226 designates quantization indices, 230 designates coded picture information, 231 designates coded block information, 232 designates quantization parameter information, and 233 designates motion vector information, all of which are output from thevariable length decoder 311 as coding parameters as the result of the decoding, and are supplied to the image codeddata synthesizer 320 as the coded data aftersignal processing 221. Thevariable length decoder 311 with such an arrangement is included in the image coded data analyzer 310 in this embodiment 11. Thequantization parameter information 232 andmotion vector information 233 are handled as one of the codedblock information 231. - The
reference numeral 324 designates a coding parameter correction/transform portion which transforms the coding parameters such as thequantization indices 226,coded picture information 230, codedblock information 231,quantization parameter information 232 andmotion vector information 233 fed from the image coded data analyzer 310 as the coded data aftersignal processing 221 so that they match the coding system after the transform. Thereference numeral 323 designates a variable length coder that carries out coding of the transformed output from the coding parameter correction/transform portion 324, and supplies the resultant second image codeddata 240 to thedecoding processor 50. The coding parameter correction/transform portion 324 andvariable length coder 323 are included in the image codeddata synthesizer 320 of the embodiment 11. - Next, the operation will be described.
- The following processing is carried out to make difference between the decoding procedure of the first image coded
data 220 input to the image codeddata analyzer 310 and that of the second image codeddata 240 output from the image codeddata synthesizer 320. For example, let us consider the case where the input first image codeddata 220 is based on theMPEG 1, and the output second image codeddata 240 is based on H.261 defining the coding system for visual telephone and video conference. In the image codeddata analyzer 310, thevariable length decoder 311 extracts from the input first image codeddata 220 based on theMPEG 1 the coding parameters such asquantization indices 226,coded picture information 230, codedblock information 231,quantization parameter information 232 andmotion vector information 233, and supplies them to the image codeddata synthesizer 320 as the coded data aftersignal processing 221. In the image codeddata synthesizer 320, the coding parameter correction/transform portion 324 receives the coded data aftersignal processing 221, and transforms the coding parameters in theMPEG 1 representation into those in the H.261 representation. - More concrete description will be provided of the transform from the
MPEG 1 image coded data to the H.261 image coded data, taking themotion vector information 233 as an example. Although {fraction (1/2)} pixel motion vectors are available in theMPEG 1, only integer multiple accuracy motion vectors are available in the H.261. Accordingly, the coding parameter correction/transform portion 324 carries out the transform in which it extracts only the integer portion of each of theMPEG 1 motion vectors which is considered optimum, and uses it as the H.261 motion vectors. Likewise, the coding parameters such as thequantization parameter information 232 andquantization indices 226 are transformed into parameter values which considered optimum in the H.261 unless they are used in exactly the same sense in both theMPEG 1 and H.261. The coding parameters transformed by the coding parameter correction/transform portion 324 such that they match the H.261 coding system are synthesized to the second image codeddata 240 by thevariable length coder 323 and is output therefrom. - Thus, according to the embodiment 11, it is not necessary to decode the first image coded
data 220 to an image once, and then re-encodes the image by an image coder based on the required coding system, thereby offering an advantage of implementing a small, low cost apparatus. This holds true even when the first image codeddata 220 input to the image codeddata analyzer 310 is transformed and output from the image codeddata synthesizer 320 as the second image codeddata 240 which is processed in the decoding procedure different from that of the first image codeddata 220. - Embodiment 12
- FIG. 11 is a block diagram showing the internal configuration of the image coded
data analyzer 310 and image codeddata synthesizer 320 of an embodiment 12 of the image codeddata re-encoding apparatus 30 in accordance with the present invention, which transforms image size between the first image codeddata 220 and the second image codeddata 240. In this figure, the same or like portions are designated by the same reference numerals as in FIG. 7, and the description thereof is omitted here. The present embodiment 12 differs from theembodiment 4 in that the image size is defined by theinformation 223 which commands the transform, and that the coefficient deletion/addition/correction portion 321 deletes or corrects thetransform coefficients 227 which are obtained through the inverse quantization by theinverse quantizer 312, by using the image size instructed by theinformation 223 commanding the transform. - Next, the operation will be described.
- Let us assume that the transform is carried out between the first image coded
data 220 which is input to the image codeddata analyzer 310 and the second image codeddata 240 which is output from the image codeddata synthesizer 320, each including an image signal of a different image size, and that thecoding processor 40 which generates the first image codeddata 220 and thedecoding processor 50 which decodes the second image codeddata 240 carry out processing based on transform coding including the motion compensative prediction, transform and quantization as shown in FIGS. 2 and 4, respectively. When the inversely quantized transform coefficients or quantization indices, which are extracted in the image codeddata analyzer 310, are increased or decreased, they are corrected in accordance with the ratio of the image sizes to be transformed. This makes it possible to reduce the substantial degradation in the resolution or unnatural images which readily occur during the transform. - Referring to FIG. 11, the present embodiment will be described in more detail.
- In the image coded data analyzer310 as shown in FIG. 11, the
variable length decoder 311 decodes the first image codeddata 220 to thequantization indices 226. Theinverse quantizer 312 carries out inverse quantization, and supplies theresultant transform coefficients 227 to the image codeddata synthesizer 320 as the coded data aftersignal processing 221. In the image codeddata synthesizer 320, the coefficient deletion/addition/correction portion 321 receives thetransform coefficients 227 together with theinformation 223 commanding the transform. The coefficient deletion/addition/correction portion 321 performs revision like deletion/addition/correction of the transform coefficients to be increased or decreased in accordance with the image sizes to be transformed which are defined by theinformation 223 commanding the transform. When reducing the image size, for example, the number of samplings of the image after transform is reduced as compared with that of the original image, and hence the frequency components of the image after transform will be lowered. Accordingly, some form of band limit is required, which is performed on thetransform coefficients 227. Thus, the revision is carried out in transforming the coded data of the original image to that of a smaller size image by suppressing high frequency components. The revised correctedtransform coefficients 228 is fed from the coefficient deletion/addition/correction portion 321 to thequantizer 322 which generates thequantization indices 229 by the quantization. Then, thevariable length coder 323 applies the variable length coding to thequantization indices 229, and supplies the resultant second image codeddata 240 to thedecoding processor 50. - Thus, according to the embodiment 12, since the transform coefficients to be increased or decreased are corrected in accordance with the image sizes to be transformed when performing the transform of the coded data involving the image size transform, an advantage is gained that it becomes possible to prevent the substantial degradation in the resolution or the occurrence of unnatural images, and the degradation in the image quality after the image size transform.
- Embodiment 13
- Although the transform coefficients to be increased or decreased are revised in accordance with the ratio of the image sizes which are transformed between the first image coded
data 220 and the second image codeddata 240 in the embodiment 12, the dimension of the motion vectors used for the motion compensation may be revised in accordance with the ratio of the image sizes to be transformed, to transform the first image codeddata 220 to the second image codeddata 240 of a different image size. This is implemented by an embodiment 13. - FIG. 12 is a block diagram showing the internal configuration of the image coded
data analyzer 310 and image codeddata synthesizer 320 of the embodiment 13 of the image codeddata re-encoding apparatus 30 in accordance with the present invention. In this figure, thereference numeral 233 designates motion vector information for motion compensation which is output from thevariable length decoder 311 in the course of variable length decoding of the first image codeddata 220 supplied from thecoding processor 40. The remaining portions are designated by the same reference numerals as in FIG. 11, and the description thereof is omitted here. The present embodiment 13 differs from the embodiment 12 in that thevariable length decoder 311 produces themotion vector information 233 besides thequantization indices 226, and that the coefficient deletion/addition/correction portion 321 carries out the revision such as the increase or decrease, or correction of thetransform coefficients 227 by using themotion vector information 233 besides the image sizes to be transformed which are defined by theinformation 223 commanding the transform. - Next, the operation will be described.
- In the image coded data analyzer310 as shown in FIG. 12, the
variable length decoder 311 decodes the first image codeddata 220, supplies theinverse quantizer 312 with theresultant quantization indices 226, and supplies the image codeddata synthesizer 320 with themotion vector information 233 which is obtained in the course of the decoding. Theinverse quantizer 312 performs the inverse quantization of thequantization indices 226, and transfers theresultant transform coefficients 227 to the image codeddata synthesizer 320 as the coded data aftersignal processing 221. In the image codeddata synthesizer 320, on the other hand, the coefficient deletion/addition/correction portion 321 receives the inversely quantizedtransform coefficients 227 andmotion vector information 233, together with theinformation 223 which commands the transform and defines the image sizes to be transformed. The coefficient deletion/addition/correction portion 321 corrects the dimension of the motion vectors in accordance with the ratio of the image sizes defined by theinformation 223 commanding the transform. Thus changing the dimension of the motion vectors in accordance with the transform ratio of the image sizes makes it possible to utilize the vectors used before the transform in substantially the same form. Accordingly, retrieval of the motion vectors based on the calculated vectors can improve the retrieval efficiency as compared with original retrieval. - Thus, according to the embodiment 13, since the motion vectors are corrected in accordance with the ratio of image sizes to be transformed, the search efficiency can be improved of the motion vectors in the image coded data after the transform. As a result, the motion compensative search in a narrower range based on the corrected motion vectors can offer a characteristic nearly equivalent to that obtained when the motion compensative search is carried out in a wider range.
- Embodiment 14
- FIG. 13 is a block diagram showing the internal configuration of the image coded
data analyzer 310 and image codeddata synthesizer 320 of an embodiment 14 of the image codeddata re-encoding apparatus 30 in accordance with the present invention, which transforms the sequence of images between the first image codeddata 220 and the second image codeddata 240, in which like portions are designated by the same reference numerals as in FIG. 12, and the description thereof is omitted here. In the present embodiment 14, the sequence of the images to be transformed is defined by theinformation 223 commanding the transform. - In the image coded data analyzer310 in FIG. 13, the
reference numeral 313 designates an inverse transformer that carries out inverse transform of the inversely quantizedtransform coefficients 227 output from theinverse quantizer 312 by applying the inverse DCT operation or the like to thetransform coefficients inverse transformer 313 to that of the motion compensator which will be described later. Thereference numeral 234 designates a decoded image output from 10 theadder 314 as the coded data aftersignal processing image 234 is stored temporarily. Thereference numeral 316 designates the motion compensator which applies the motion compensation to the decoded image one cycle before, which is read out of theframe memory 315, on the basis of themotion vectors 233 generated in the decoding process by thevariable length decoder 311, and supplies the processing result to theadder 314. - In the image coded
data synthesizer 320, thereference numeral 325 designates an inverse quantizer for carrying out inverse quantization of thequantization indices 229 output from thequantizer inverse quantizer 325 by applying the inverse DCT or the like to this output, and 327 designates an adder for adding the processing result of theinverse transformer 326 to a frame image output from a motion searcher which will be described later. Thereference numeral 328 designates a frame memory which stores the output of theadder 327 temporarily, and 329 designates a motion searcher. Themotion searcher 329, receiving the image data one cycle before which is read out of theframe memory 328, themotion vector information 233 from the image codeddata analyzer 310, the decodedimage 234 as the coded data aftersignal processing 221, and thesignal 223 which commands the transform and defines the sequence and information of the images to be transformed, estimates the dimension of the motion vectors extracted in response to the sequence information of the images to be transformed, and carries out the motion search on the basis of the estimation result. Thereference numeral 235 designates frame images delivered from themotion searcher 329 to theadder 327 and asubtracter 330 for calculating the difference between the decodedimage 234 which is supplied from the image coded data analyzer 310 as the coded data aftersignal processing 221 and theimage data 235 output from themotion searcher 329. Thereference numeral 331 designates a transformer for applying the transform processing such as DCT operation or the like to the output of thesubtracter 330, and for supplying the processing result to thequantizer 322. - Next, the operation will be described.
- Let us assume that the sequences are transformed of the image signals included in the first image coded
data 220 which is input to the image codeddata analyzer 310 and in the second image codeddata 240 which is output from the image codeddata synthesizer 320, and that thecoding processor 40 which generates the first image codeddata 220 and thedecoding processor 50 which decodes the second image codeddata 240 carry out processing based on transform coding including the motion compensative prediction, transform and quantization as shown in FIGS. 2 and 4, respectively. In this case, the image codeddata analyzer 310 extracts the motion vectors used in the motion compensation, and themotion searcher 329 estimates the S dimension of the motion vectors in response to the sequence information of the images to be transformed. Performing such a search based on the estimated motion vectors when transforming the picture type to be coded, for example, enables the motion search in a narrower range involving a large volume of calculations to keep efficiency equivalent to that of the motion compensative search in a wider range. - In the image coded data analyzer310 as shown in FIG. 13, the
variable length decoder 311 decodes the first image codeddata 220, and supplies theresultant quantization indices 226 to theinverse quantizer 312, and themotion vector information 233 obtained in the decoding process to the image codeddata synthesizer 320 and themotion compensator 316. Theinverse quantizer 312 inversely quantizes thequantization indices 226, and theinverse transformer 313 applies the inverse transform such as inverse DCT operation to theresultant transform coefficients 227 to supply its result to theadder 314. On the other hand, themotion compensator 316 carries out the motion compensation of the decoded image one cycle before which is output from theframe memory 315 in accordance with themotion vector information 233 fed from thevariable length decoder 311, and provides the result to theadder 314. Theadder 314 adds the frame images fed from theinverse transformer 313 and themotion compensator 316 to generate the decodedimage 234, and supplies it to the image codeddata synthesizer 320 as the coded data aftersignal processing 221. Thus, the image coded data analyzer 310 not only performs equivalent operation to the decoding by a common transform coding system with a motion compensation, but also supplies the image codeddata synthesizer 320 with the decodedimage 234 as the coded data aftersignal processing 221 and themotion vector information 233 extracted by thevariable length decoder 311. - In the image coded
data synthesizer 320, on the other hand, themotion searcher 329 receives the decodedimage 234 and themotion vector information 233, together with theinformation 223 which commands the transform and defines the sequence information of the images to be transformed. Themotion searcher 329 also receives the image data one cycle before which is read out of theframe memory 328. This image data is generated by inversely quantizing in theinverse quantizer 325 thequantization indices 229 output from thequantizer 322, by applying the inverse transform such as the inverse DCT to the output of theinverse quantizer 325 in theinverse transformer 326, by adding by theadder 327 the processing results in theinverse transformer 326 and themotion searcher 329, and by storing the addition result. Themotion searcher 329 not only generates theframe images 235 on the basis of these data to achieve operation equivalent to that of the common transform coding system with motion compensation, but also estimates the motion vectors of the current coded data by using themotion vector information 233 fed from the image codeddata analyzer 310, and the sequence information about the images to be transformed, which is defined by theinformation 223 commanding the transform, thereby carrying out the motion search based on the estimation result. Theframe image 235 output from themotion searcher 329 is supplied to thesubtracter 330 which calculates the difference between theframe image 235 and the decodedimage 234 delivered from the image coded data analyzer 310 as the coded data aftersignal processing 221. Then, thetransformer 331 transforms the output of thesubtracter 330 by applying the DCT operation thereto. Thequantizer 322 performs the quantization, and thevariable length coder 323 carries out coding of the resultant transform coefficients to the second image codeddata 240. Thus, the image codeddata synthesizer 320 not only performs equivalent operation to the coding in the common transform coding system with the motion compensation by inputting the decodedimage 234 andmotion vector information 233 from the image codeddata analyzer 310, but also estimates in themotion searcher 329 the dimension of the motion vectors extracted in the image coded data analyzer 310 in response to the sequence information of the images to be changed which is provided by theinformation 223 commanding the transform, thereby performing the motion search based on the estimation result. - Thus, according to the embodiment 14, the delay involved in coding is shortened by performing transform which changes the sequence of the images at reproduction and that of the images after the transform. Furthermore, it can improve the efficiency of coding using the transformed motion vectors by estimating the dimension of the motion vectors in response to the sequence information about the images to be transformed in the processing for changing the sequence of the images to be transformed.
- Embodiment 15
- FIG. 14 is a block diagram showing the internal configuration of the image coded
data analyzer 310 and image codeddata synthesizer 320 of an embodiment 15 of the image codeddata re-encoding apparatus 30 in accordance with the present invention, which transforms the number of frames of the image signal included in the first image codeddata 220 and that of the image signal included in the second image codeddata 240, in which the like portions are designated by the same reference numerals as in FIG. 8, and the description thereof is omitted here. The present embodiment 15 differs from theembodiment 8 in that theinformation 223 commanding the transform defines the frame rate of the images, and that the coefficient deletion/addition/correction portion 321 deletes or corrects thetransform coefficients 227, which are inversely quantized by theinverse quantizer 312, by using the frame rate information of the images based on theinformation 223 commanding the transform. - Next, the operation will be described.
- Let us assume that the transform is carried out between the first image coded
data 220 and the second image codeddata 240, and that the number of frames associated with the first image codeddata 220 differ from that associated with the second image codeddata 240, in which each number of frames is defined as that of the decoded image signal per unit time when the first image codeddata 220 input to the image codeddata analyzer 310 and the second image codeddata 240 output from the image codeddata synthesizer 320 are decoded into images. When an image coding mode, which is used as information extracted from the first image codeddata 220, indicates the image which will not be used for future coding of images, the data of the image is deleted. This enables the transform of the frame rate to be achieved without having substantial effect on the quality of the future images. - In the image coded data analyzer310 as shown in FIG. 14, the
variable length decoder 311 decodes the first image codeddata 220, and supplies theinverse quantizer 312 with theresultant quantization indices 226, and the image codeddata synthesizer 320 with thecoded picture information 230 which is obtained in the course of decoding. Theinverse quantizer 312 carries out the inverse quantization of thequantization indices 226, and sends theresultant transform coefficients 227 to the image codeddata synthesizer 320 as the coded data aftersignal processing 221. In the image codeddata synthesizer 320, on the other hand, the coefficient deletion/addition/correction portion 321 receives the inversely quantizedtransform coefficients 227 and thecoded picture information 230 together with theinformation 223 which commands the transform and defines the image frame rate, and decides whether or not these input data indicate the coded picture which will not be used for the prediction in the future coding. If so, the data of the picture is deleted by applying the method of theembodiment 8, thereby achieving the transform of the frame rate equivalently. - Thus, according to the embodiment 15, an advantage is gained that the transform is readily achieved between the television signals of different systems by varying the number of frames between the image signal included in the first image coded
data 220 before the transform and that included in the second image codeddata 240 after the transform. - Embodiment 16
- FIG. 15 is a block diagram showing the internal configuration of the image coded
data analyzer 310 and image codeddata synthesizer 320 of an embodiment 16 of the image codeddata re-encoding apparatus 30 in accordance with the present invention, which estimates the quantization parameters from the decoded images of the first image codeddata 220, and used them for the quantization for generating the second image codeddata 240, in which the like portions are designated by the same reference numerals as in FIG. 13, and the description thereof is omitted here. In this figure, thereference numeral 236 designates a decoded image theinverse transformer 313 decodes by applying the inverse DCT operation to thetransform coefficients 227 output from theinverse quantizer 312. The decodedimage 236 is sent to the image codeddata synthesizer 320 as the coded data aftersignal processing 221. Thereference numeral 332 designates a quantization estimator which estimates the quantization parameters during the generation of the first image codeddata 220 from the decodedimage 236 fed as the coded data aftersignal processing quantization estimator 332 to thequantizer transformer 331 generates from the decodedimage 236. - Next, the operation will be described.
- When transforming the rate of images between the first image coded
data 220 and the second image codeddata 240, the image codeddata analyzer 310 extracts from the first image codeddata 220 the quantization parameters in the quantization process, and the image codeddata synthesizer 320 carries out the quantization using the quantization parameters. This means that the image codeddata analyzer 310 performs the decoding and the image codeddata synthesizer 320 carries out the coding. This makes it possible to achieve the highest quality transform when the bit rate and the coding system are each identical before and after the transform. In addition, even if the bit rates differ, optimum transform can be achieved by controlling the quantization parameters in accordance with the ratio of the bit rates. - In the image coded data analyzer310 as shown in FIG. 15, the
variable length decoder 311 carries out the variable length decoding of the first image codeddata 220 to generate thequantization indices 226, and theinverse quantizer 312 inversely quantizes them to generate thetransform coefficients 227. Theinverse transformer 313 applies the inverse DCT operation to thetransform coefficients 227 to generate the decodedimage 236, and transfers it to the image codeddata synthesizer 320 as the coded data aftersignal processing 221. Thus, the image codeddata analyzer 310 performs operation equivalent to common decoding, thereby outputting the decodedimage 236 as the coded data aftersignal processing 221. In the image codeddata synthesizer 320, on the other hand, thetransformer 331 receives the decodedimage 236 sent as the coded data aftersignal processing 221, and applies the DCT operation to it to generate thetransform coefficients 238 which are input to thequantizer 322. Thequantization estimator 332 also receives the decodedimage 236, estimates the quantization parameters during the generation of the first image codeddata 220, and supplies the resultantquantization parameter information 237 to thequantizer 322. Thequantizer 322 quantizes thetransform coefficients 238 fed from thetransformer 331 in response to thequantization parameter information 237 from thequantization estimator 332. Then, thevariable length coder 323 carries out coding of thequantization indices 229 to generate and output the second image codeddata 240. - Thus, according to the embodiment 16, estimating the quantization parameters from the decoded
image 236 enables the image quality associated with the newly generated second image codeddata 240 to be improved, thereby offering an advantage of keeping the image quality after the transform in relay transmission. In particular, when the rates before and after the transform are identical, using the estimated quantization parameters enable the image quality to be kept even after repeating a plurality of coding operations. In addition, even if the rates are different before and after the transform, an advantage is gained that optimum transform can be achieved by controlling the quantization parameters in accordance with the ratio of the rates. - Embodiment 17
- The foregoing embodiments can offer configuration not only for forming communication images, but also for multi-site image transform systems, or systems for copying image data in storage media.
- Furthermore, although the portions playing a major role in the transform, such as the coefficient deletion/addition/
correction portion 321, coding parameter correction/transform portion 324 andquantization estimator 332 are placed in the image codeddata synthesizer 320, it is not necessary that they are provided in the image codeddata synthesizer 320. For example, they may be placed in the image coded data analyzer 310 or in outside independently.
Claims (21)
1. An image coded data re-encoding apparatus which receives a first image coded data generated by a coding processor performing coding of a digital input image signal, and which generates a second image coded data by performing a digital signal processing on said first image coded data, said image coded data re-encoding apparatus comprising:
an image coded data analyzer for generating coded data after signal processing by performing a first digital signal processing on said first image coded data; and
an image coded data synthesizer for generating said second image coded data by performing on said coded data after signal processing a second digital signal processing based on multiple signals associated with said first image coded data by using said coded data after signal processing output from said image coded data analyzer and said multiple signals.
2. The image coded data re-encoding apparatus as claimed in claim 1 , wherein
said image coded data analyzer extracts said multiple signals in the course of generating said coded data after signal processing by performing said first digital signal processing on said first image coded data, and
said image coded data synthesizer generates said second image coded data by performing said second digital signal processing on said coded data after signal processing based on said multiple signals by using said coded data after signal processing and said multiple signals which are output from said image coded data analyzer.
3. The image coded data re-encoding apparatus as claimed in claim 1 , wherein
said image coded data re-encoding apparatus further comprises a separator for separating from said first image coded data said multiple signals which have been externally combined with said first image coded data and cannot be extracted from said first image coded data in said first digital signal processing for generating said coded data after signal processing, and
said image coded data synthesizer generates said second image coded data by performing on said coded data after signal processing said second digital signal processing based on said multiple signals by using said coded data after signal processing output from said image coded data analyzer and said multiple signals output from said separator.
4. The image coded data re-encoding apparatus as claimed in claim 1 , wherein
said image coded data analyzer decodes said first image coded data by performing said first digital signal processing on said first image coded data to generate decoded image data as said coded data after signal processing,
said image coded data re-encoding apparatus further comprises an information extractor/estimator for extracting or estimating said multiple signals needed for said second digital signal processing from said coded data after signal processing generated by said image coded data analyzer, and
said image coded data synthesizer generates said second image coded data by performing on said coded data after signal processing said second digital signal processing based on said multiple signals by using said coded data after signal processing output from said image coded data analyzer and said multiple signals output from said information extractor/estimator.
5. The image coded data re-encoding apparatus as claimed in any one of claims 1-4, wherein said image coded data synthesizer generates said second image coded data with a data amount different from a data amount of said first image coded data input to said image coded data analyzer, when said image coded data synthesizer generates said second image coded data by performing said second digital signal processing on said coded data after signal processing from said image coded data analyzer in response to said multiple signals from said image coded data analyzer.
6. The image coded data re-encoding apparatus as claimed in claim 5 ,
wherein said image coded data analyzer extracts transform coefficients or quantization indices, which are inversely quantized, and
wherein said image coded data re-encoding apparatus further comprises a coefficient deletion/addition/correction portion for deleting part of said transform coefficients or quantization indices, which are extracted by said image coded data analyzer, and for correcting said transform coefficients or quantization indices in accordance with a ratio of amounts of data to be transformed.
7. The image coded data re-encoding apparatus as claimed in claim 5 ,
wherein said image coded data analyzer extracts transform coefficients or quantization indices, which are inversely quantized, and
wherein said image coded data re-encoding apparatus further comprises a coefficient deletion/addition/correction portion for deleting part of said transform coefficients or quantization indices, which are extracted by said image coded data analyzer, and are weighted in accordance with relationships between said transform coefficients or quantization indices and their neighboring transform coefficients or quantization indices.
8. The image coded data re-encoding apparatus as claimed in claim 5 ,
wherein said image coded data analyzer extracts transform coefficients or quantization indices, which are inversely quantized, and
wherein said image coded data re-encoding apparatus further comprises a coefficient deletion/addition/correction portion for adding, to said transform coefficients or quantization indices which are extracted by said image coded data analyzer, new transform coefficients or quantization indices after correcting said new transform coefficients or quantization indices in accordance with a ratio of amounts of data to be transformed.
9. The image coded data re-encoding apparatus as claimed in claim 5 ,
wherein said image coded data analyzer extracts transform coefficients or quantization indices, which are inversely quantized, and
wherein said image coded data re-encoding apparatus further comprises a coefficient deletion/addition/correction portion for adding, to said transform coefficients or quantization indices which are extracted by said image coded data analyzer, new transform coefficients or quantization indices after predicting transform coefficients or quantization indices including said new transform coefficients or quantization indices and their neighboring transform coefficients or quantization indices.
10. The image coded data re-encoding apparatus as claimed in claim 5 ,
wherein said image coded data analyzer extracts transform coefficients or quantization indices, which are inversely quantized, and generates a coding parameter designating a picture type of a current image to be processed, and
wherein said image coded data re-encoding apparatus further comprises a coefficient deletion/addition/correction portion for increasing a ratio of deletion of said transform coefficients or quantization indices, which are extracted by said image coded data analyzer, if said coding parameter designating the picture type indicates, when decision is made whether or not said current image to be processed is used for prediction in future coding, that said picture type is not used for the prediction in the future coding.
11. The image coded data re-encoding apparatus as claimed in claim 5 ,
wherein said image coded data analyzer extracts transform coefficients or quantization indices, which are inversely quantized, and generates a coding parameter designating a picture type of a current image to be processed, and
wherein said image coded data re-encoding apparatus further comprises a coefficient deletion/addition/correction portion for decreasing a ratio of deletion of said transform coefficients or quantization indices, which are extracted by said image coded data analyzer, if said coding parameter designating the picture type indicates, when decision is made whether or not said current image to be processed is used for prediction in future coding, that said picture type is used for the prediction in the future coding.
12. The image coded data re-encoding apparatus as claimed in claim 5 ,
wherein said image coded data analyzer extracts transform coefficients or quantization indices, which are inversely quantized, and generates a coding parameter designating a picture type of a current image to be processed, and a coding parameter designating a predictive type of an image block of the current image, and
wherein said image coded data re-encoding apparatus further comprises a coefficient deletion/addition/correction portion for increasing a ratio of deletion of said transform coefficients or quantization indices, which are extracted by said image coded data analyzer, if said coding parameter designating the predictive type of said image block indicates, when decision is made whether a current image to be processed is used for prediction in future coding by using the coding parameters generated by said image coded data analyzer, that said image block is not used for the prediction in the future coding, even if said coding parameter designating the picture type indicates that the picture type is used for the prediction in the future coding.
13. The image coded data re-encoding apparatus as claimed in any one of claims 1-4, wherein said image coded data synthesizer generates said second image coded data whose decoding procedure differs from a decoding procedure of said first image coded data input to said image coded data analyzer, when said image coded data synthesizer generates said second image coded data by performing said second digital signal processing on said coded data after signal processing from said image coded data analyzer in response to said multiple signals from said image coded data analyzer.
14. The image coded data re-encoding apparatus as claimed in claim 13 , wherein
said image coded data analyzer extracts from said first image coded data various types of coding parameters with an expression form in said decoding procedure of said first image coded data, and
said image coded data re-encoding apparatus further comprises a coding parameter corrector/transformer for correcting and transforming said expression form of said various types of coding parameters, which are extracted by said image coded data analyzer, from said expression form in said decoding procedure of said first image coded data to an expression form in the decoding procedure of said second image coded data.
15. The image coded data re-encoding apparatus as claimed in any one of claims 1-4, wherein said image coded data synthesizer generates said second image coded data including an image signal whose image size differs in time or space from an image size of an image signal included in said first image coded data input to said image coded data analyzer, when said image coded data synthesizer generates said second image coded data by performing said second digital signal processing on said coded data after signal processing from said image coded data analyzer in response to said multiple signals from said image coded data analyzer.
16. The image coded data re-encoding apparatus as claimed in claim 15 ,
wherein said image coded data analyzer extracts transform coefficients or quantization indices, which are inversely quantized, and
wherein said image coded data re-encoding apparatus further comprises a coefficient deletion/addition/correction portion for changing an amount of said transform coefficients or quantization indices extracted by said image coded data analyzer, and for correcting said transform coefficients or quantization indices, which are extracted by said image coded data analyzer, in accordance with a ratio of the image sizes to be transformed.
17. The image coded data re-encoding apparatus as claimed in claim 15 ,
wherein said image coded data analyzer extracts motion vectors used for motion compensation, and
wherein said image coded data re-encoding apparatus further comprises a coefficient deletion/addition/correction portion for correcting dimension of said motion vectors extracted by said image coded data analyzer in accordance with a ratio of the image sizes to be transformed.
18. The image coded data re-encoding apparatus as claimed in any one of claims 1-4, wherein said image coded data synthesizer generates said second image coded data including an image signal whose sequence differs from a sequence of an image signal included in said first image coded data input to said image coded data analyzer, when said image coded data synthesizer generates said second image coded data by performing said second digital signal processing on said coded data after signal processing from said image coded data analyzer in response to said multiple signals from said image coded data analyzer.
19. The image coded data re-encoding apparatus as claimed in claim 18 ,
wherein said image coded data analyzer extracts motion vectors used for motion compensation, and
wherein said image coded data re-encoding apparatus further comprises a motion searcher for estimating dimension of said motion vectors extracted by said image coded data analyzer in accordance with said sequence of the image signals to be transformed.
20. The image coded data re-encoding apparatus as claimed in any one of claims 1-4, wherein said image coded data synthesizer generates said second image coded data whose decoded image signal includes a number of frames per unit time different from a number of frames per unit time of a decoded image signal of said first image coded data input to said image coded data analyzer, when said image coded data synthesizer generates said second image coded data by performing said second digital signal processing on said coded data after signal processing from said image coded data analyzer in response to said multiple signals from said image coded data analyzer.
21. The image coded data re-encoding apparatus as claimed in claim 5 ,
wherein said image coded data analyzer generates as the coded data after signal processing a decoded image obtained by decoding said first image coded data,
wherein said image coded data re-encoding apparatus further comprises a quantization estimator for estimating, from said decoded image output from said image coded data analyzer, quantization parameters obtained in the course of generating said first image coded data, and
wherein said image coded data synthesizer generates said second image coded data by using said quantization parameters estimated by said quantization estimator.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
US09/250,404 US20030133510A1 (en) | 1996-08-05 | 1999-02-16 | Image coded data re-encoding apparatus |
Applications Claiming Priority (4)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
JP206226/96 | 1996-08-05 | ||
JP20622696A JPH1051766A (en) | 1996-08-05 | 1996-08-05 | Image coding data converter |
US80623797A | 1997-02-24 | 1997-02-24 | |
US09/250,404 US20030133510A1 (en) | 1996-08-05 | 1999-02-16 | Image coded data re-encoding apparatus |
Related Parent Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
US80623797A Division | 1996-08-05 | 1997-02-24 |
Publications (1)
Publication Number | Publication Date |
---|---|
US20030133510A1 true US20030133510A1 (en) | 2003-07-17 |
Family
ID=16519859
Family Applications (4)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
US09/250,404 Abandoned US20030133510A1 (en) | 1996-08-05 | 1999-02-16 | Image coded data re-encoding apparatus |
US09/351,281 Expired - Fee Related US6442207B1 (en) | 1996-08-05 | 1999-07-12 | Image coded data re-encoding apparatus without once decoding the original image coded data |
US09/351,283 Expired - Fee Related US6246438B1 (en) | 1996-08-05 | 1999-07-12 | Image coded data re-encoding apparatus without once decoding the original image coded data |
US09/351,282 Expired - Fee Related US6222887B1 (en) | 1996-08-05 | 1999-07-12 | Image coded data re-encoding apparatus without once decoding the original image coded data |
Family Applications After (3)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
US09/351,281 Expired - Fee Related US6442207B1 (en) | 1996-08-05 | 1999-07-12 | Image coded data re-encoding apparatus without once decoding the original image coded data |
US09/351,283 Expired - Fee Related US6246438B1 (en) | 1996-08-05 | 1999-07-12 | Image coded data re-encoding apparatus without once decoding the original image coded data |
US09/351,282 Expired - Fee Related US6222887B1 (en) | 1996-08-05 | 1999-07-12 | Image coded data re-encoding apparatus without once decoding the original image coded data |
Country Status (9)
Country | Link |
---|---|
US (4) | US20030133510A1 (en) |
EP (1) | EP0823822B1 (en) |
JP (1) | JPH1051766A (en) |
KR (2) | KR19980018014A (en) |
CN (2) | CN1183488C (en) |
AU (1) | AU692408B2 (en) |
CA (1) | CA2197820C (en) |
DE (1) | DE69739548D1 (en) |
TW (1) | TW324813B (en) |
Cited By (8)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US20020181584A1 (en) * | 2001-03-29 | 2002-12-05 | Patrice Alexandre | Method and device for controlling the quality of video data |
US20050240769A1 (en) * | 2004-04-22 | 2005-10-27 | Gassoway Paul A | Methods and systems for computer security |
US20060224646A1 (en) * | 2005-03-30 | 2006-10-05 | Chhavi Kishore | System (s), method (s), and apparatus for converting unsigned fixed length codes (decoded from exponential golomb codes) to signed fixed length codes |
US20070147496A1 (en) * | 2005-12-23 | 2007-06-28 | Bhaskar Sherigar | Hardware implementation of programmable controls for inverse quantizing with a plurality of standards |
US20080181296A1 (en) * | 2007-01-16 | 2008-07-31 | Dihong Tian | Per multi-block partition breakpoint determining for hybrid variable length coding |
CN101998051A (en) * | 2009-08-07 | 2011-03-30 | 三洋电机株式会社 | Image display control device, imaging device provided with the image display control device, image processing device |
US20150341652A1 (en) * | 2014-05-22 | 2015-11-26 | Microsoft Corporation | Re-encoding image sets using frequency-domain differences |
US10404993B2 (en) | 2015-03-10 | 2019-09-03 | Huawei Technologies Co., Ltd. | Picture prediction method and related apparatus |
Families Citing this family (54)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US5715009A (en) | 1994-03-29 | 1998-02-03 | Sony Corporation | Picture signal transmitting method and apparatus |
GB2327577B (en) * | 1997-07-18 | 2002-09-11 | British Broadcasting Corp | Re-encoding decoded signals |
JP2008109701A (en) * | 1997-11-05 | 2008-05-08 | Sony Corp | Method and device for converting digital signal |
JP4513856B2 (en) * | 1997-11-05 | 2010-07-28 | ソニー株式会社 | Digital signal conversion method and digital signal conversion apparatus |
KR100592651B1 (en) * | 1997-11-27 | 2006-06-23 | 브리티쉬 텔리커뮤니케이션즈 파블릭 리미티드 캄퍼니 | Transcoding |
US6151361A (en) * | 1997-12-29 | 2000-11-21 | Thomson Licensing S.A. | System and method for reducing multipass signal degradation |
US6574274B2 (en) * | 1998-02-27 | 2003-06-03 | Sony Corporation | Picture signal processing system, decoder, picture signal processing method, and decoding method |
CA2265089C (en) * | 1998-03-10 | 2007-07-10 | Sony Corporation | Transcoding system using encoding history information |
US6483947B1 (en) | 1998-03-17 | 2002-11-19 | Matsushita Electric Industrial Co., Ltd. | Video signal processing apparatus |
KR20010013152A (en) * | 1998-03-31 | 2001-02-26 | 요트.게.아. 롤페즈 | Modifying data which has been coded |
JP4200541B2 (en) * | 1998-04-03 | 2008-12-24 | ソニー株式会社 | Data conversion apparatus and method, signal recording / reproducing apparatus, and reproducing apparatus and method |
US6459731B1 (en) * | 1998-08-28 | 2002-10-01 | Lucent Technologies Inc. | Technique for video communications using a coding-matched filter arrangement |
KR100571687B1 (en) * | 1999-02-09 | 2006-04-18 | 소니 가부시끼 가이샤 | Coding system and its method, coding device and its method, decoding device and its method, recording device and its method, and reproducing device and its method |
JP2000244921A (en) * | 1999-02-24 | 2000-09-08 | Matsushita Electric Ind Co Ltd | Method and device for coding video image |
JP2000278688A (en) * | 1999-03-24 | 2000-10-06 | Sony Corp | Motion vector detector, its method and image processor |
TW455821B (en) * | 1999-09-03 | 2001-09-21 | Ind Tech Res Inst | Signal process method and device |
JP3694888B2 (en) * | 1999-12-03 | 2005-09-14 | ソニー株式会社 | Decoding device and method, encoding device and method, information processing device and method, and recording medium |
WO2001058167A1 (en) * | 2000-02-04 | 2001-08-09 | Koninklijke Philips Electronics N.V. | Quantization method for bit rate transcoding applications |
JP4378824B2 (en) * | 2000-02-22 | 2009-12-09 | ソニー株式会社 | Image processing apparatus and method |
FR2809573B1 (en) | 2000-05-26 | 2002-08-16 | Thomson Broadcast Systems | METHOD FOR ENCODING A VIDEO IMAGE STREAM |
KR100386583B1 (en) * | 2000-11-30 | 2003-06-02 | 엘지전자 주식회사 | Apparatus and method for transcoding video |
US6756921B2 (en) * | 2000-12-27 | 2004-06-29 | Mitsubishi Denki Kabushiki Kaisha | Multiple quality data creation encoder, multiple quality data creation decoder, multiple quantity data encoding decoding system, multiple quality data creation encoding method, multiple quality data creation decoding method, and multiple quality data creation encoding/decoding method |
US7082220B2 (en) * | 2001-01-25 | 2006-07-25 | Sony Corporation | Data processing apparatus |
US7155010B2 (en) * | 2001-05-04 | 2006-12-26 | Hewlett-Packard Development Company, L.P. | Signal format that facilitates easy scalability of encrypted streams |
US7054362B1 (en) | 2001-05-29 | 2006-05-30 | Cisco Technology, Inc. | Methods and apparatus for updating a reduction ratio |
WO2002100112A1 (en) * | 2001-06-03 | 2002-12-12 | Seelive Ltd. | System and method for rapid video compression |
WO2003001695A1 (en) * | 2001-06-26 | 2003-01-03 | Cineform, Inc. | Method and apparatus for real-time editing of plural content streams |
JP2003116104A (en) * | 2001-10-02 | 2003-04-18 | Sony Corp | Information processing apparatus and information processing method |
JP4189952B2 (en) * | 2002-11-22 | 2008-12-03 | 株式会社リコー | Encoding / decoding device, encoding / decoding program, and storage medium |
SG140441A1 (en) * | 2003-03-17 | 2008-03-28 | St Microelectronics Asia | Decoder and method of decoding using pseudo two pass decoding and one pass encoding |
US9171577B1 (en) | 2003-04-25 | 2015-10-27 | Gopro, Inc. | Encoding and decoding selectively retrievable representations of video content |
JP4196726B2 (en) * | 2003-05-14 | 2008-12-17 | ソニー株式会社 | Image processing apparatus, image processing method, recording medium, and program |
US7889792B2 (en) | 2003-12-24 | 2011-02-15 | Apple Inc. | Method and system for video encoding using a variable number of B frames |
EP1553779A1 (en) * | 2004-01-12 | 2005-07-13 | Deutsche Thomson-Brandt GmbH | Data reduction of video streams by selection of frames and partial deletion of transform coefficients |
US7869503B2 (en) * | 2004-02-06 | 2011-01-11 | Apple Inc. | Rate and quality controller for H.264/AVC video coder and scene analyzer therefor |
US7492820B2 (en) | 2004-02-06 | 2009-02-17 | Apple Inc. | Rate control for video coder employing adaptive linear regression bits modeling |
US7986731B2 (en) * | 2004-02-06 | 2011-07-26 | Apple Inc. | H.264/AVC coder incorporating rate and quality controller |
JP4243205B2 (en) * | 2004-02-18 | 2009-03-25 | 日本放送協会 | Moving picture coding apparatus, method and program thereof |
US7551752B2 (en) * | 2004-04-26 | 2009-06-23 | Graphic Security Systems Corporation | Systems and methods for authenticating objects using multiple-level image encoding and decoding |
US7773815B2 (en) * | 2004-07-14 | 2010-08-10 | Hewlett-Packard Development Company, L.P. | System and method for compressing compressed data |
JP2006094415A (en) * | 2004-09-27 | 2006-04-06 | Toshiba Corp | Video image apparatus and video image streaming method |
EP1815687A1 (en) * | 2004-11-16 | 2007-08-08 | Koninklijke Philips Electronics N.V. | Sender side channel adaptive video transmission |
US8014597B1 (en) | 2006-03-22 | 2011-09-06 | Woodman Labs | Method for efficient compression and decoding of single sensor color image data |
US7796825B2 (en) * | 2007-02-16 | 2010-09-14 | Cisco Technology, Inc. | Losslessly improving compression of compressed image data |
US8055087B2 (en) * | 2008-04-29 | 2011-11-08 | Sony Corporation | Sample level variation for spatial sub-sampled images |
US8059909B2 (en) * | 2008-04-29 | 2011-11-15 | Sony Corporation | Adaptive generation of irregular spatial sub-sampling for images |
US8059908B2 (en) * | 2008-04-29 | 2011-11-15 | Sony Corporation | Adaptive area of influence filter for irregular spatial sub-sampled images |
JP4656452B2 (en) * | 2008-05-20 | 2011-03-23 | ソニー株式会社 | Image signal processing device, image signal processing method, prediction coefficient generation device, prediction coefficient generation processing method, and program for causing computer to execute each method |
WO2010086021A1 (en) * | 2009-01-30 | 2010-08-05 | Telefonaktiebolaget L M Ericsson (Publ) | Method and apparatus for efficient downstream video processing based on upstream metric processing and provision |
US8755444B2 (en) * | 2010-08-06 | 2014-06-17 | Qualcomm Incorporated | Two-stage entropy decoding |
SG189354A1 (en) | 2010-10-11 | 2013-05-31 | Graphic Security Systems Corp | Method for constructing a composite image incorporating a hidden authentication image |
EP2751990A4 (en) | 2011-08-29 | 2015-04-22 | Icvt Ltd | Controlling a video content system |
US9344720B2 (en) * | 2012-08-07 | 2016-05-17 | Apple Inc. | Entropy coding techniques and protocol to support parallel processing with low latency |
US20140328406A1 (en) * | 2013-05-01 | 2014-11-06 | Raymond John Westwater | Method and Apparatus to Perform Optimal Visually-Weighed Quantization of Time-Varying Visual Sequences in Transform Space |
Family Cites Families (16)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
JPH02179186A (en) | 1988-12-29 | 1990-07-12 | Toshiba Corp | Video conference system |
DE69224227T2 (en) * | 1991-04-18 | 1998-07-09 | Ampex | Method and arrangement for determining a quantization factor for processes with data compression and decompression |
JP3163830B2 (en) * | 1993-03-29 | 2001-05-08 | ソニー株式会社 | Image signal transmission method and apparatus |
KR0134343B1 (en) | 1993-04-13 | 1998-04-29 | 김광호 | Coding device and method of quantization level |
JP3085024B2 (en) | 1993-06-01 | 2000-09-04 | 松下電器産業株式会社 | Image recompressor and image recording device |
US5446491A (en) | 1993-12-21 | 1995-08-29 | Hitachi, Ltd. | Multi-point video conference system wherein each terminal comprises a shared frame memory to store information from other terminals |
US5537440A (en) | 1994-01-07 | 1996-07-16 | Motorola, Inc. | Efficient transcoding device and method |
JPH07212240A (en) | 1994-01-26 | 1995-08-11 | Hitachi Ltd | Information conversion device |
US5541852A (en) | 1994-04-14 | 1996-07-30 | Motorola, Inc. | Device, method and system for variable bit-rate packet video communications |
JPH07336681A (en) | 1994-06-08 | 1995-12-22 | Matsushita Electric Ind Co Ltd | Picture converting device |
EP0687112B1 (en) * | 1994-06-08 | 2006-09-20 | Matsushita Electric Industrial Co., Ltd. | Image conversion apparatus |
DE69522861T2 (en) | 1994-06-30 | 2002-04-11 | Koninkl Philips Electronics Nv | Method and device for code conversion of coded data stream |
JP3058028B2 (en) * | 1994-10-31 | 2000-07-04 | 三菱電機株式会社 | Image encoded data re-encoding device |
US5600646A (en) * | 1995-01-27 | 1997-02-04 | Videoserver, Inc. | Video teleconferencing system with digital transcoding |
JPH0984011A (en) | 1995-09-14 | 1997-03-28 | Hitachi Ltd | Moving image coding system converter |
JPH0993133A (en) | 1995-09-28 | 1997-04-04 | Nippon Telegr & Teleph Corp <Ntt> | Coding form converter |
-
1996
- 1996-08-05 JP JP20622696A patent/JPH1051766A/en active Pending
- 1996-12-31 TW TW085116380A patent/TW324813B/en not_active IP Right Cessation
-
1997
- 1997-02-05 KR KR1019970003607A patent/KR19980018014A/en active Search and Examination
- 1997-02-17 AU AU14718/97A patent/AU692408B2/en not_active Ceased
- 1997-02-18 CA CA 2197820 patent/CA2197820C/en not_active Expired - Fee Related
- 1997-02-25 EP EP19970103019 patent/EP0823822B1/en not_active Expired - Lifetime
- 1997-02-25 DE DE69739548T patent/DE69739548D1/en not_active Expired - Lifetime
- 1997-04-03 CN CNB971102333A patent/CN1183488C/en not_active Expired - Fee Related
-
1999
- 1999-02-16 US US09/250,404 patent/US20030133510A1/en not_active Abandoned
- 1999-07-12 US US09/351,281 patent/US6442207B1/en not_active Expired - Fee Related
- 1999-07-12 US US09/351,283 patent/US6246438B1/en not_active Expired - Fee Related
- 1999-07-12 US US09/351,282 patent/US6222887B1/en not_active Expired - Fee Related
-
2000
- 2000-08-25 KR KR1020000049540A patent/KR100316492B1/en not_active IP Right Cessation
-
2001
- 2001-09-19 CN CNB011408502A patent/CN1211761C/en not_active Expired - Fee Related
Cited By (14)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US20020181584A1 (en) * | 2001-03-29 | 2002-12-05 | Patrice Alexandre | Method and device for controlling the quality of video data |
US7082167B2 (en) * | 2001-03-29 | 2006-07-25 | Nextream S.A. | Method and device for controlling the quality of video data |
US20050240769A1 (en) * | 2004-04-22 | 2005-10-27 | Gassoway Paul A | Methods and systems for computer security |
US7801935B2 (en) * | 2005-03-30 | 2010-09-21 | Broadcom Corporation | System (s), method (s), and apparatus for converting unsigned fixed length codes (decoded from exponential golomb codes) to signed fixed length codes |
US20060224646A1 (en) * | 2005-03-30 | 2006-10-05 | Chhavi Kishore | System (s), method (s), and apparatus for converting unsigned fixed length codes (decoded from exponential golomb codes) to signed fixed length codes |
US20070147496A1 (en) * | 2005-12-23 | 2007-06-28 | Bhaskar Sherigar | Hardware implementation of programmable controls for inverse quantizing with a plurality of standards |
US20080181296A1 (en) * | 2007-01-16 | 2008-07-31 | Dihong Tian | Per multi-block partition breakpoint determining for hybrid variable length coding |
US8126062B2 (en) * | 2007-01-16 | 2012-02-28 | Cisco Technology, Inc. | Per multi-block partition breakpoint determining for hybrid variable length coding |
CN101998051A (en) * | 2009-08-07 | 2011-03-30 | 三洋电机株式会社 | Image display control device, imaging device provided with the image display control device, image processing device |
US20150341652A1 (en) * | 2014-05-22 | 2015-11-26 | Microsoft Corporation | Re-encoding image sets using frequency-domain differences |
US9554145B2 (en) * | 2014-05-22 | 2017-01-24 | Microsoft Technology Licensing, Llc | Re-encoding image sets using frequency-domain differences |
US10404993B2 (en) | 2015-03-10 | 2019-09-03 | Huawei Technologies Co., Ltd. | Picture prediction method and related apparatus |
US10659803B2 (en) | 2015-03-10 | 2020-05-19 | Huawei Technologies Co., Ltd. | Picture prediction method and related apparatus |
US11178419B2 (en) | 2015-03-10 | 2021-11-16 | Huawei Technologies Co., Ltd. | Picture prediction method and related apparatus |
Also Published As
Publication number | Publication date |
---|---|
US6222887B1 (en) | 2001-04-24 |
CA2197820C (en) | 2001-02-13 |
CN1172998A (en) | 1998-02-11 |
AU1471897A (en) | 1998-02-12 |
EP0823822B1 (en) | 2009-08-26 |
CA2197820A1 (en) | 1998-02-06 |
EP0823822A2 (en) | 1998-02-11 |
US6246438B1 (en) | 2001-06-12 |
KR100316492B1 (en) | 2001-12-12 |
CN1211761C (en) | 2005-07-20 |
US6442207B1 (en) | 2002-08-27 |
DE69739548D1 (en) | 2009-10-08 |
CN1183488C (en) | 2005-01-05 |
TW324813B (en) | 1998-01-11 |
KR19980018014A (en) | 1998-06-05 |
AU692408B2 (en) | 1998-06-04 |
EP0823822A3 (en) | 1999-05-06 |
CN1347063A (en) | 2002-05-01 |
JPH1051766A (en) | 1998-02-20 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
US6442207B1 (en) | Image coded data re-encoding apparatus without once decoding the original image coded data | |
EP0710030B1 (en) | Image coded data re-encoding apparatus | |
US5537440A (en) | Efficient transcoding device and method | |
US5764298A (en) | Digital data transcoder with relaxed internal decoder/coder interface frame jitter requirements | |
US5260782A (en) | Adaptive DCT/DPCM video signal coding method | |
US6434197B1 (en) | Multi-functional transcoder for compressed bit streams | |
US6289052B1 (en) | Methods and apparatus for motion estimation using causal templates | |
US5315326A (en) | Efficient coding/decoding apparatuses for processing digital image signal | |
US6546052B1 (en) | Image processing apparatus and method, and computer-readable memory | |
KR20020042166A (en) | Apparatus and method for transcoding video | |
EP0956704B1 (en) | Overhead data processor in a memory efficient image processing system | |
JPH07288804A (en) | Re-coding device for picture signal | |
US6980598B2 (en) | Programmable vertical filter for video encoding | |
JP2000224587A (en) | System and device for encoding picture | |
Moiron et al. | H. 264/AVC to MPEG-2 video transcoding architecture | |
US20060140274A1 (en) | Transcoder and method used therein | |
EP1104198A1 (en) | MPEG transcoder | |
KR980012955A (en) | Image encoder and / or decoder using iterative motion prediction / compensation | |
JP4875894B2 (en) | Image coding apparatus and image coding method | |
JP2002218470A (en) | Method for converting image encoded data rate and device for converting image encoding rate | |
JP2003092762A (en) | Image encoding data converter | |
JP3770466B2 (en) | Image coding rate conversion apparatus and image coding rate conversion method | |
WO2004023820A1 (en) | A device and a system for providing a change of a video signal | |
JPH0730895A (en) | Picture processor and its processing method |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
STCB | Information on status: application discontinuation |
Free format text: ABANDONED -- FAILURE TO RESPOND TO AN OFFICE ACTION |