CA2275563C - Moving picture decoding apparatus and moving picture decoding method - Google Patents
Moving picture decoding apparatus and moving picture decoding method Download PDFInfo
- Publication number
- CA2275563C CA2275563C CA 2275563 CA2275563A CA2275563C CA 2275563 C CA2275563 C CA 2275563C CA 2275563 CA2275563 CA 2275563 CA 2275563 A CA2275563 A CA 2275563A CA 2275563 C CA2275563 C CA 2275563C
- Authority
- CA
- Canada
- Prior art keywords
- bit number
- compression
- moving picture
- picture decoding
- quantization
- 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.)
- Expired - Fee Related
Links
- 238000000034 method Methods 0.000 title claims description 7
- 238000007906 compression Methods 0.000 claims abstract description 85
- 230000006835 compression Effects 0.000 claims abstract description 85
- 238000003079 width control Methods 0.000 claims abstract description 25
- 238000013139 quantization Methods 0.000 claims description 56
- 230000003044 adaptive effect Effects 0.000 description 2
- 238000010586 diagram Methods 0.000 description 2
- 206010039897 Sedation Diseases 0.000 description 1
- 238000006243 chemical reaction Methods 0.000 description 1
- 238000013144 data compression Methods 0.000 description 1
- 230000003247 decreasing effect Effects 0.000 description 1
- 239000000284 extract Substances 0.000 description 1
- 230000036280 sedation Effects 0.000 description 1
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
- H04N19/42—Methods or arrangements for coding, decoding, compressing or decompressing digital video signals characterised by implementation details or hardware specially adapted for video compression or decompression, e.g. dedicated software implementation
- H04N19/423—Methods or arrangements for coding, decoding, compressing or decompressing digital video signals characterised by implementation details or hardware specially adapted for video compression or decompression, e.g. dedicated software implementation characterised by memory arrangements
-
- 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
- H04N19/61—Methods or arrangements for coding, decoding, compressing or decompressing digital video signals using transform coding in combination with predictive coding
Abstract
A memory compression and expansion section (11 of Fig.
1) applies compression to a decoded image, and it is stored in a frame memory. Based on an occupied content of the frame memory (Fig. 1 106), in case that the coded bit number for a single or a plurality of memory compression processing units or for every control unit of memory compression processing exceeds the bit number of a memory access unit or is lacking, a memory access width control section (110 of Fig. 1) applies control to a quantizer control section (109 of Fig. 1) so that the coded bit number is conformed to be equal to or less than the bit number of the memory access unit.
1) applies compression to a decoded image, and it is stored in a frame memory. Based on an occupied content of the frame memory (Fig. 1 106), in case that the coded bit number for a single or a plurality of memory compression processing units or for every control unit of memory compression processing exceeds the bit number of a memory access unit or is lacking, a memory access width control section (110 of Fig. 1) applies control to a quantizer control section (109 of Fig. 1) so that the coded bit number is conformed to be equal to or less than the bit number of the memory access unit.
Description
MOVING PICTURE DECODING APPARATUS AND MOVING PICTURE
DECODING METHOD
BACKGROUND OF THE INVENTION
The present invention relates to a moving picture decoding apparatus, and especially, to a moving picture decoding apparatus capable of reducing a capacity of a frame memory when coded moving pictures are decoded.
Conventionally, this kind of a moving picture decoding apparatus is used as a method for reducing a frame memory capacity. As the conventional moving picture decoding apparatus, in JP-A-261635/1997 for example, as an arrangement for reducing a frame memory capacity necessary for storage of reference images and re-ordering decoded images, an arrangement is disclosed, which comprises a data compression circuit for compressing a decoded image and a data expansion circuit for expanding a compressed data. Also, in JP-A-247673/1997, an arrangement is proposed, in which a moving pictures data of which data content has been reduced by data content reducing means for reducing a data content of a decoded moving pictures data is stored in a frame memory, and a data content of a data read from the frame memory is restored by data content restoring means.
Also, in JP-A-4550/1998 for example, an apparatus is proposed, in which memory reduction is facilitated using Adaptive Differential Pulse Code Modulation (ADPCM). In the apparatus described in the publication, compression is conducted by estimating a change of each~block of n Xm . pixels compressed using orthogonal translation, encoding a first pixel of each block with p bits in accordance with the ADPCM, encoding an estimated value of the change with nXh bits, and further, encoding a difference value between other each pixel following the first pixel and an l0 average value of all pixels in the blocks with p-k bits.
However, in the conventional moving picture decoding apparatus proposed in the above-described JP-A-4550/1998 and so forth, of which objective is to reduce capacity of the frame memory, there is a task that efficient access to the memory cannot be realized.
The reason thereof is that an access width of the memory is not at all taken into account in a memory~compression and expansion section of the above=described moving picture decoding apparatus.
SDI~iARY OF THE INVENTION
An aspect of the present invention is to solve the above-mentioned.
Also, an aspect thereof is to provide a moving picture decoding apparatus for reducing capacity of a frame memory necessary when an image is decoded by means of compression operation and accomplishing compression and expansion processing of a memory content with a high speed, and further, for decoding an image with high quality.
According to one broad aspect of the invention, a moving picture decoding apparatus to which a compressed stream generated using an inter-frame prediction system is input, said apparatus comprises:
compression means for compressing a decoded image when the signal is stored in storage means; expansion means for expanding a compressed image stored in said storage means; quantization control means for controlling quantization on compression in said compression means;
and access width control means for controlling said quantization control means so that bit allocation control is conducted so as to be in conformity with the bit number of an access unit of said storage means.
In the moving picture decoding apparatus of the present invention, said access width control means comprises means for controlling said quantization control means so that the coded bit number for a single or a plurality of compression processing units or for every control unit of compression processing is in conformity with the bit number of an access unit of raid storage means in case that the coded bit number exceeds the bit number of an access unit of said storage means or is lacking.
Furthermore, in the moving picture decoding apparatus of the present invention, said compression means and expansion means conduct compression and expansion, respectively, in accordance with a pixel difference prediction encoding system.
Furthermore, the moving picture decoding apparatus of the present invention further comprising a plurality of quantizers and a plurality of quantization characteristic tables.
Furthermore, the moving picture decoding apparatus of the present invention further comprising a plurality of quantizers and a quantization characteristic table being shared by said plurality of quantizers.
Furthermore, in the moving picture decoding apparatus of the present invention, said compression means and expansion means conduct compression and expansion, respectively, in accordance with an orthogonal translation encoding system.
Furthermore, in the moving picture decoding apparatus of the present invention, said access width control means conducts control using information included in the compressed stream.
_ 5 _ Furthermore, in the moving picture decoding apparatus of the present invention, said storage means is a frame memory.
According to another broad aspect of the invention, there is a moving picture decoding apparatus to which a compressed stream generated using an inter-frame prediction system is input, said apparatus comprises:
compression means for compressing a decoded image; storage means for storing a compressed image in said compression means; expansion means for expanding the compressed image stored in said storage means; quantization control means for controlling quantization on compression in said compression means; and an access width control means for applying bit allocation control to said quantization control means so as to be in conformity with the bit number of an access unit of said storage means, and said quantization control means controls quantization in said compression means based on access width information from said access width control means so that generated information content for one or a plurality of compression processing units or for every control unit of compression processing is equal to or less than the bit number of an access unit of said storage means in case that the generated information content exceeds the bit number of an access unit of said storage means or is lacking.
According to a further broad aspect of the invention, there is a moving picture decoding method comprising the steps of: detecting the coded bit number for one or a plurality of compression processing units or for every control unit of compression processing and controlling said coded bit number so that said coded bit number is in conformity with the bit number of an access unit of a storage means when said detected coded bit number exceeds - 5a -the bit number of an access unit of said storage means or is lacking.
BRIEF DESCRIPTION OF THE DRAWINGS
This and other objects, features and advantages of the present invention will become more apparent upon a reading of the following detailed description and drawings, in which:
Fig. 1 is a block diagram showing an arrangement of a moving picture decoding apparatus of the first embodiment of the present invention;
Fig. 2 is a view for explaining the principle of the present invention and a view for showing an access bit width and a width of a quantized data;
Fig. 3 is a view for explaining the principle of the present invention and a view for showing an access bit width and a width of a quantized data;
Fig. 4 is a view for showing an arrangement of a memory compression and expansion section in the first embodiment of the present invention and Fig. 5 is a block diagram showing an arrangement of the second embodiment of the present invention.
DESCRIPTION OF THE EMBODIMENTS
A form of working of the present invention will be explained. In a preferable form of working of a moving picture decoding apparatus ir.~ the present invention, compression is applied to a dlecoded image in a memory compressor (105 of Fig. 1), amd it is stored in a frame memory (106 of Fig. 1). BasedL on an occupied content of the frame memory (106 of Fig. 1), in case that the coded data bit number for a single or a plurality of memory compression processing units or for every control unit of memory compression processing' exceeds the bit number of a memory access unit or is lacking, a memory access width control section (110 of Fig. 1) applies control to a quantizer control section (109 of Fig. 1) so that the coded data bit number is conformed to be equal to or less than the bit number of the memory access unit.
First, the principle of the present invention will be explained. A decoded image is compressed using various compression methods. However, the coded data bit number per compression processing unit, which is allocated by compression operation, is allocated based on a compression ratio, and the bit number of a memory access unit is not taken into account.
It is assuming that the bit number of the memory access unit is M1 bits, the coded data bit numbers of a compression processing unit, which is allocated based on a compression ratio, are N1, N2, N3, "' Nn, respectively, 7 _ and M1 < N1 + N2 + "' + Nn. In this case, a coded data of n blocks cannot be extracted through one access.
Accordingly, by decreasing the bit number allocated by N1 + N2 + "' + Nn by N1 + N2 + "' + Nn - M1, an. allocated bit can be made equal to or less than M1, and therefore, the coded data can be taken out through one access.
An example of n = 3 as the n blocks is shown in Fig. 2.
By subtracting more than or equal to M2 (M2 = (N1 + N2 +
N3) - M1) bits from N1 + N2 + N3 bits, an allocated bit can be made equal to or less than M1.
On the other hand, as shown in Fag. 3, in case that the allocated bit is less than the bit number M1 of a memory access unit, and a frame memory has capacity, allocation of the bit number of the coded data can be also increased.
In the present invention, on the basis of such a principal, the memory access width control section (110 of Fig. 1) applies control of an access width of the frame memory to the quantization control section.
In a form of working of the present invention, the memory compressor and memory expander may :have an arrangement in which compression and expansion are conducted, respectively, in accordance with DPCM.
In a form of working of the present invention, based on access width information from the access width control section, in case that a generated information content. for _ $ _ a single or a plurality of memory compression processing units or for every control unit of memory compression processing exceeds the bit number of a memory access unit, quantization control is conducted by preparing a plurality of quantizers having differer.~t quantization characteristics, and a quanti.zation table, and by selecting a quantizer that is. adapted to an access width.
In this case, an arrangement may be adopted, in which one quantization table is shared by the plurality of quantizers having different g,,uantization characteristics.
In other words, by variably changing a reference interval of one quantization table in accordance with an obtained quantization characteristic (resolution and so forth), it is possible to obtain the different quantization characteristics while one quantization table is shared.
Also, in a form of working of the present invention, the memory compressor and the memory expander may conduct compression and expansion, respectively, in accordance with an orthogonal translation encoding system.
Moreover, in a form of working of the present invention, the access width control section may apply access width control to the quantization control section using information included in the compressed stream. Below, a detailed explanation will be :made in accordance with an embodiment.
The first embodiment of the; present invention will be explained in detail by referring to drawings. Fig. 1 is a view showing an arrangement of one embodiment of the present invention.
Referring to Fig. 1, this embodiment is constructed by comprising a variable length decoder 101, an inverse quantizer 102, an inverse discrete cosine transducer (Inverse DCT) 103, an adder 104, a motion compensation section 108, and a memory compression and expansion section 11. The memory compression and expansion section 11 is constructed by comprising a memory compressor 105, a frame memory 106, a memory expander 107, a quantizer control section 109, and a memory width control section 110.
A compressed moving pictures stream encoded by a system such an ISOIS13818-2 (MPEG-2VIDE0) is supplied as an input to the variable length decoder 101.
The variable length decoder 101 performs variable length decoding, and supplies a result thereof to the inverse quantizer 102.
The inverse quantizer 102 performs inverse quantization, and supplies a result thereof to the inverse discrete cosine transducer 103.
The inverse discrete cosine transducer 103 transduces a conversion coefficient into a:n image, and supplies a result thereof to the adder x.04.
The adder 104 performs addition of the image supplied from the inverse discrete cosine transducer 103 and a predicted image supplies from the motion compensation section 108, and supplies a result thereof to the memory compression and expansion section 11.
The memory compression and expansion section 11 applies compression and expansion operation to a decoded image, and supplies it to the motion. compensation section 108.
The motion compensation section 108 performs motion compensation based on a motion vector supplied from the variable length decoder 101 and the image supplied from the memory compression and expansion section 11, and supplies the predicted image to the adder 104.
The memory compressor 105 of the memory compression and expansion section 11 applies compression to the image supplied from the adder 104, and stores it in the frame memory 106.
The memory expander 107 extracts the compressed image from the frame memory 106 and expands it, and supplies it to the motion compensation section 108.
The quantization control section 109 applies control for quantization of the image to the memory compressor 105 and the memory expander 107.
The access width control se~;,tion 110 applies control to the quantization control section 109 using information of the frame memory 106 so that an information content for a single or a plurality of memory compression processing units or for every control unit of memory compression processing is conformed to be. equal to or less than the bit number of a memory access unit.
As the compression encoding in the memory compressor and the memory expander, pixel difference encoding, adaptive pixel difference encoding, DCT translation, Hadamard translation, and Wavelet translation and so forth, for example, are used.
Fig. 4 is a view showing one example of an arrangement of the memory compression and expansion section 11 in one embodiment of the present invention in case that the memory compressor 105 and the memory expander 107 are constructed of the pixel difference encoding. The memory compressor 105 and the memory expander 107 in Fig. 1 correspond to a memory compressor 21 and a memory expander 23 in Fig. 4.
Referring to Fig. 4, the memory compressor 21 is constructed by comprising a subtracter 201, a quantizer 202, a fixed-length decoder 203, an inverse quantizer 204, an adder 205, and a predictor 206. The memory expander 23 is constructed by comprising .a fixed-length decoder 207, an inverse quantizer 208, an .adder 209, and a predictor 210.
An image to be compressed is supplied to the subtracter 201 as an input, and the subt:racter 201 performs subtraction of this input signal and a predicted value supplied from the predictor x;06, and supplies a prediction error to the quantizer 202.
The quantizer 202 performs quantization of the prediction error in accordance with the quantization control section 24, and supplies it to the fixed-length decoder 203 and the inverse g:uantizer 204.
The fixed-length decoder 203 encodes the quantized value and stores it in the frame memory 22. In the inverse quantizer 204, the adder 205 and the predictor 206, inverse quantization and local decoding are performed, and an output from the predictor 206 is supplied to the subtracter 201.
The memory access width control section 25 takes account of a compression ratio based on an occupied ratio of the frame memory 22, and applies bit allocation control to the quantization control section 24, which is in conformity with the bit number of a memory access unit.
The quantization control sedation 24 performs quantization control in which memory access width information is added, and applies control to the quantizer 202 so that an information content generated for a single or a plurality of memory compression processing units or for every control unit of memory compression processing is to be equal to or less than t:he bit number of a memory access unit.
In this embodiment, if the compression processing unit is assumed to be constant, a block of a single pixel or a plurality of m X n pixels becomes to be a compression processing unit.
The memory expander 23 receives a compressed image from the frame memory 22, and performs expansion of the compressed image by applying expansion and inverse quantization thereto, and outputs the image.
Next, an example will be explained, in which one memory access width control is realized. One quantizer having a fixed compression ratio is used for quantization of a prediction error. In such a case, in case that an information content generated for a single or a plurality of memory compression processing units or for every control unit of memory compression processing exceeds the bit number of a memory access unit, a plurality of quantizers are prepared, and control is conducted so that the information content is in conformity with an access width. In this example, since an occupied content of the frame memory can be fixed, in:Eormation from the frame memory is not required. In addition, an arrangement can be adopted, in which a pluralit5~ of quantizers having quantization characteristics different from each other share one quantization characteristic table.
On the other hand, in a case of orthogonal translation and so forth, it is difficult: to fix an information content generated for a single or a plurality of memory compression processing units or for every control unit of memory compression processing'. In such a case, use of an occupied content of the frame memory is convenient.
Next, the second embodiment of the present invention will be explained. In the above-described first embodiment, the access width control section 110 conducts control of the quantizer control section 109 using the access memory width and the occupied content of the frame memory.
Fig. 5 is a view showing an arrangement of the second embodiment of the present invention. Referring to Fig. 5, in the second embodiment of t:he present invention, the access width control section 110 receives information from the inverse quantizer 102, and applies control to the quantization control section 109, in which information included in a compressed stream is added, in other words, in which information during quantization is added.
Accordingly, by taking account of the information of the compressed stream and a relat:Lon between positions on an image, and so forth, allocation of the information content can be realized, which makes image quality be improved.
As explained above, in accordance with the present invention, advantages are effected as described below.
The first advantage of the present invention is that compression and expansion processing for the contents of the memory can be conducted with a high speed.
The reason thereof is that, in the present invention, since quantization control which is to be equal to or less than the number of bits of a memory access unit is conducted, the number of a memory access to the compressed image necessary for the expansion is reduced.
The second advantage of the present invention is that, by conducting allocation control of an information content using the information included in the compressed stream, an error of an pixel can be suppressed and image quality can be improved.
The reason thereof is that, in the present invention, by taking account of the information of the compressed stream and a relation between positi~~ns on an image, and so forth, allocation of the information content can be realized, which makes image quality be :improved.
DECODING METHOD
BACKGROUND OF THE INVENTION
The present invention relates to a moving picture decoding apparatus, and especially, to a moving picture decoding apparatus capable of reducing a capacity of a frame memory when coded moving pictures are decoded.
Conventionally, this kind of a moving picture decoding apparatus is used as a method for reducing a frame memory capacity. As the conventional moving picture decoding apparatus, in JP-A-261635/1997 for example, as an arrangement for reducing a frame memory capacity necessary for storage of reference images and re-ordering decoded images, an arrangement is disclosed, which comprises a data compression circuit for compressing a decoded image and a data expansion circuit for expanding a compressed data. Also, in JP-A-247673/1997, an arrangement is proposed, in which a moving pictures data of which data content has been reduced by data content reducing means for reducing a data content of a decoded moving pictures data is stored in a frame memory, and a data content of a data read from the frame memory is restored by data content restoring means.
Also, in JP-A-4550/1998 for example, an apparatus is proposed, in which memory reduction is facilitated using Adaptive Differential Pulse Code Modulation (ADPCM). In the apparatus described in the publication, compression is conducted by estimating a change of each~block of n Xm . pixels compressed using orthogonal translation, encoding a first pixel of each block with p bits in accordance with the ADPCM, encoding an estimated value of the change with nXh bits, and further, encoding a difference value between other each pixel following the first pixel and an l0 average value of all pixels in the blocks with p-k bits.
However, in the conventional moving picture decoding apparatus proposed in the above-described JP-A-4550/1998 and so forth, of which objective is to reduce capacity of the frame memory, there is a task that efficient access to the memory cannot be realized.
The reason thereof is that an access width of the memory is not at all taken into account in a memory~compression and expansion section of the above=described moving picture decoding apparatus.
SDI~iARY OF THE INVENTION
An aspect of the present invention is to solve the above-mentioned.
Also, an aspect thereof is to provide a moving picture decoding apparatus for reducing capacity of a frame memory necessary when an image is decoded by means of compression operation and accomplishing compression and expansion processing of a memory content with a high speed, and further, for decoding an image with high quality.
According to one broad aspect of the invention, a moving picture decoding apparatus to which a compressed stream generated using an inter-frame prediction system is input, said apparatus comprises:
compression means for compressing a decoded image when the signal is stored in storage means; expansion means for expanding a compressed image stored in said storage means; quantization control means for controlling quantization on compression in said compression means;
and access width control means for controlling said quantization control means so that bit allocation control is conducted so as to be in conformity with the bit number of an access unit of said storage means.
In the moving picture decoding apparatus of the present invention, said access width control means comprises means for controlling said quantization control means so that the coded bit number for a single or a plurality of compression processing units or for every control unit of compression processing is in conformity with the bit number of an access unit of raid storage means in case that the coded bit number exceeds the bit number of an access unit of said storage means or is lacking.
Furthermore, in the moving picture decoding apparatus of the present invention, said compression means and expansion means conduct compression and expansion, respectively, in accordance with a pixel difference prediction encoding system.
Furthermore, the moving picture decoding apparatus of the present invention further comprising a plurality of quantizers and a plurality of quantization characteristic tables.
Furthermore, the moving picture decoding apparatus of the present invention further comprising a plurality of quantizers and a quantization characteristic table being shared by said plurality of quantizers.
Furthermore, in the moving picture decoding apparatus of the present invention, said compression means and expansion means conduct compression and expansion, respectively, in accordance with an orthogonal translation encoding system.
Furthermore, in the moving picture decoding apparatus of the present invention, said access width control means conducts control using information included in the compressed stream.
_ 5 _ Furthermore, in the moving picture decoding apparatus of the present invention, said storage means is a frame memory.
According to another broad aspect of the invention, there is a moving picture decoding apparatus to which a compressed stream generated using an inter-frame prediction system is input, said apparatus comprises:
compression means for compressing a decoded image; storage means for storing a compressed image in said compression means; expansion means for expanding the compressed image stored in said storage means; quantization control means for controlling quantization on compression in said compression means; and an access width control means for applying bit allocation control to said quantization control means so as to be in conformity with the bit number of an access unit of said storage means, and said quantization control means controls quantization in said compression means based on access width information from said access width control means so that generated information content for one or a plurality of compression processing units or for every control unit of compression processing is equal to or less than the bit number of an access unit of said storage means in case that the generated information content exceeds the bit number of an access unit of said storage means or is lacking.
According to a further broad aspect of the invention, there is a moving picture decoding method comprising the steps of: detecting the coded bit number for one or a plurality of compression processing units or for every control unit of compression processing and controlling said coded bit number so that said coded bit number is in conformity with the bit number of an access unit of a storage means when said detected coded bit number exceeds - 5a -the bit number of an access unit of said storage means or is lacking.
BRIEF DESCRIPTION OF THE DRAWINGS
This and other objects, features and advantages of the present invention will become more apparent upon a reading of the following detailed description and drawings, in which:
Fig. 1 is a block diagram showing an arrangement of a moving picture decoding apparatus of the first embodiment of the present invention;
Fig. 2 is a view for explaining the principle of the present invention and a view for showing an access bit width and a width of a quantized data;
Fig. 3 is a view for explaining the principle of the present invention and a view for showing an access bit width and a width of a quantized data;
Fig. 4 is a view for showing an arrangement of a memory compression and expansion section in the first embodiment of the present invention and Fig. 5 is a block diagram showing an arrangement of the second embodiment of the present invention.
DESCRIPTION OF THE EMBODIMENTS
A form of working of the present invention will be explained. In a preferable form of working of a moving picture decoding apparatus ir.~ the present invention, compression is applied to a dlecoded image in a memory compressor (105 of Fig. 1), amd it is stored in a frame memory (106 of Fig. 1). BasedL on an occupied content of the frame memory (106 of Fig. 1), in case that the coded data bit number for a single or a plurality of memory compression processing units or for every control unit of memory compression processing' exceeds the bit number of a memory access unit or is lacking, a memory access width control section (110 of Fig. 1) applies control to a quantizer control section (109 of Fig. 1) so that the coded data bit number is conformed to be equal to or less than the bit number of the memory access unit.
First, the principle of the present invention will be explained. A decoded image is compressed using various compression methods. However, the coded data bit number per compression processing unit, which is allocated by compression operation, is allocated based on a compression ratio, and the bit number of a memory access unit is not taken into account.
It is assuming that the bit number of the memory access unit is M1 bits, the coded data bit numbers of a compression processing unit, which is allocated based on a compression ratio, are N1, N2, N3, "' Nn, respectively, 7 _ and M1 < N1 + N2 + "' + Nn. In this case, a coded data of n blocks cannot be extracted through one access.
Accordingly, by decreasing the bit number allocated by N1 + N2 + "' + Nn by N1 + N2 + "' + Nn - M1, an. allocated bit can be made equal to or less than M1, and therefore, the coded data can be taken out through one access.
An example of n = 3 as the n blocks is shown in Fig. 2.
By subtracting more than or equal to M2 (M2 = (N1 + N2 +
N3) - M1) bits from N1 + N2 + N3 bits, an allocated bit can be made equal to or less than M1.
On the other hand, as shown in Fag. 3, in case that the allocated bit is less than the bit number M1 of a memory access unit, and a frame memory has capacity, allocation of the bit number of the coded data can be also increased.
In the present invention, on the basis of such a principal, the memory access width control section (110 of Fig. 1) applies control of an access width of the frame memory to the quantization control section.
In a form of working of the present invention, the memory compressor and memory expander may :have an arrangement in which compression and expansion are conducted, respectively, in accordance with DPCM.
In a form of working of the present invention, based on access width information from the access width control section, in case that a generated information content. for _ $ _ a single or a plurality of memory compression processing units or for every control unit of memory compression processing exceeds the bit number of a memory access unit, quantization control is conducted by preparing a plurality of quantizers having differer.~t quantization characteristics, and a quanti.zation table, and by selecting a quantizer that is. adapted to an access width.
In this case, an arrangement may be adopted, in which one quantization table is shared by the plurality of quantizers having different g,,uantization characteristics.
In other words, by variably changing a reference interval of one quantization table in accordance with an obtained quantization characteristic (resolution and so forth), it is possible to obtain the different quantization characteristics while one quantization table is shared.
Also, in a form of working of the present invention, the memory compressor and the memory expander may conduct compression and expansion, respectively, in accordance with an orthogonal translation encoding system.
Moreover, in a form of working of the present invention, the access width control section may apply access width control to the quantization control section using information included in the compressed stream. Below, a detailed explanation will be :made in accordance with an embodiment.
The first embodiment of the; present invention will be explained in detail by referring to drawings. Fig. 1 is a view showing an arrangement of one embodiment of the present invention.
Referring to Fig. 1, this embodiment is constructed by comprising a variable length decoder 101, an inverse quantizer 102, an inverse discrete cosine transducer (Inverse DCT) 103, an adder 104, a motion compensation section 108, and a memory compression and expansion section 11. The memory compression and expansion section 11 is constructed by comprising a memory compressor 105, a frame memory 106, a memory expander 107, a quantizer control section 109, and a memory width control section 110.
A compressed moving pictures stream encoded by a system such an ISOIS13818-2 (MPEG-2VIDE0) is supplied as an input to the variable length decoder 101.
The variable length decoder 101 performs variable length decoding, and supplies a result thereof to the inverse quantizer 102.
The inverse quantizer 102 performs inverse quantization, and supplies a result thereof to the inverse discrete cosine transducer 103.
The inverse discrete cosine transducer 103 transduces a conversion coefficient into a:n image, and supplies a result thereof to the adder x.04.
The adder 104 performs addition of the image supplied from the inverse discrete cosine transducer 103 and a predicted image supplies from the motion compensation section 108, and supplies a result thereof to the memory compression and expansion section 11.
The memory compression and expansion section 11 applies compression and expansion operation to a decoded image, and supplies it to the motion. compensation section 108.
The motion compensation section 108 performs motion compensation based on a motion vector supplied from the variable length decoder 101 and the image supplied from the memory compression and expansion section 11, and supplies the predicted image to the adder 104.
The memory compressor 105 of the memory compression and expansion section 11 applies compression to the image supplied from the adder 104, and stores it in the frame memory 106.
The memory expander 107 extracts the compressed image from the frame memory 106 and expands it, and supplies it to the motion compensation section 108.
The quantization control section 109 applies control for quantization of the image to the memory compressor 105 and the memory expander 107.
The access width control se~;,tion 110 applies control to the quantization control section 109 using information of the frame memory 106 so that an information content for a single or a plurality of memory compression processing units or for every control unit of memory compression processing is conformed to be. equal to or less than the bit number of a memory access unit.
As the compression encoding in the memory compressor and the memory expander, pixel difference encoding, adaptive pixel difference encoding, DCT translation, Hadamard translation, and Wavelet translation and so forth, for example, are used.
Fig. 4 is a view showing one example of an arrangement of the memory compression and expansion section 11 in one embodiment of the present invention in case that the memory compressor 105 and the memory expander 107 are constructed of the pixel difference encoding. The memory compressor 105 and the memory expander 107 in Fig. 1 correspond to a memory compressor 21 and a memory expander 23 in Fig. 4.
Referring to Fig. 4, the memory compressor 21 is constructed by comprising a subtracter 201, a quantizer 202, a fixed-length decoder 203, an inverse quantizer 204, an adder 205, and a predictor 206. The memory expander 23 is constructed by comprising .a fixed-length decoder 207, an inverse quantizer 208, an .adder 209, and a predictor 210.
An image to be compressed is supplied to the subtracter 201 as an input, and the subt:racter 201 performs subtraction of this input signal and a predicted value supplied from the predictor x;06, and supplies a prediction error to the quantizer 202.
The quantizer 202 performs quantization of the prediction error in accordance with the quantization control section 24, and supplies it to the fixed-length decoder 203 and the inverse g:uantizer 204.
The fixed-length decoder 203 encodes the quantized value and stores it in the frame memory 22. In the inverse quantizer 204, the adder 205 and the predictor 206, inverse quantization and local decoding are performed, and an output from the predictor 206 is supplied to the subtracter 201.
The memory access width control section 25 takes account of a compression ratio based on an occupied ratio of the frame memory 22, and applies bit allocation control to the quantization control section 24, which is in conformity with the bit number of a memory access unit.
The quantization control sedation 24 performs quantization control in which memory access width information is added, and applies control to the quantizer 202 so that an information content generated for a single or a plurality of memory compression processing units or for every control unit of memory compression processing is to be equal to or less than t:he bit number of a memory access unit.
In this embodiment, if the compression processing unit is assumed to be constant, a block of a single pixel or a plurality of m X n pixels becomes to be a compression processing unit.
The memory expander 23 receives a compressed image from the frame memory 22, and performs expansion of the compressed image by applying expansion and inverse quantization thereto, and outputs the image.
Next, an example will be explained, in which one memory access width control is realized. One quantizer having a fixed compression ratio is used for quantization of a prediction error. In such a case, in case that an information content generated for a single or a plurality of memory compression processing units or for every control unit of memory compression processing exceeds the bit number of a memory access unit, a plurality of quantizers are prepared, and control is conducted so that the information content is in conformity with an access width. In this example, since an occupied content of the frame memory can be fixed, in:Eormation from the frame memory is not required. In addition, an arrangement can be adopted, in which a pluralit5~ of quantizers having quantization characteristics different from each other share one quantization characteristic table.
On the other hand, in a case of orthogonal translation and so forth, it is difficult: to fix an information content generated for a single or a plurality of memory compression processing units or for every control unit of memory compression processing'. In such a case, use of an occupied content of the frame memory is convenient.
Next, the second embodiment of the present invention will be explained. In the above-described first embodiment, the access width control section 110 conducts control of the quantizer control section 109 using the access memory width and the occupied content of the frame memory.
Fig. 5 is a view showing an arrangement of the second embodiment of the present invention. Referring to Fig. 5, in the second embodiment of t:he present invention, the access width control section 110 receives information from the inverse quantizer 102, and applies control to the quantization control section 109, in which information included in a compressed stream is added, in other words, in which information during quantization is added.
Accordingly, by taking account of the information of the compressed stream and a relat:Lon between positions on an image, and so forth, allocation of the information content can be realized, which makes image quality be improved.
As explained above, in accordance with the present invention, advantages are effected as described below.
The first advantage of the present invention is that compression and expansion processing for the contents of the memory can be conducted with a high speed.
The reason thereof is that, in the present invention, since quantization control which is to be equal to or less than the number of bits of a memory access unit is conducted, the number of a memory access to the compressed image necessary for the expansion is reduced.
The second advantage of the present invention is that, by conducting allocation control of an information content using the information included in the compressed stream, an error of an pixel can be suppressed and image quality can be improved.
The reason thereof is that, in the present invention, by taking account of the information of the compressed stream and a relation between positi~~ns on an image, and so forth, allocation of the information content can be realized, which makes image quality be :improved.
Claims (9)
1 A moving picture decoding apparatus to which a compressed stream generated using an inter-frame prediction system is input, said apparatus comprises:
compression means for compressing a decoded image when the signal is stored in storage means;
expansion means for expanding a compressed image stored in said storage means;
quantization control means for controlling quantization on compression in said compression means; and access width control means for controlling said quantization control means so that bit allocation control is conducted so as to be in conformity with the bit number of an access unit of said storage means.
compression means for compressing a decoded image when the signal is stored in storage means;
expansion means for expanding a compressed image stored in said storage means;
quantization control means for controlling quantization on compression in said compression means; and access width control means for controlling said quantization control means so that bit allocation control is conducted so as to be in conformity with the bit number of an access unit of said storage means.
2 A moving picture decoding apparatus according to claim 1, wherein said access width control means comprises means for controlling said quantization control means so that the coded bit number for one or a plurality of compression processing units or for every control unit of compression processing is in conformity with the bit number of an access unit of said storage means in case that the coded bit number exceeds the bit number of an access unit of said storage means or is lacking.
3 A moving picture decoding apparatus according to claim 1, wherein said compression means and expansion means conduct compression and expansion, respectively, in accordance with a pixel difference prediction encoding system.
4 A moving picture decoding apparatus according to claim 1, further comprising a plurality of quantizers and a plurality of quantization characteristic tables.
5 A moving picture decoding apparatus according to claim 1, further comprising a plurality of quantizers and a quantization characteristic table being shared by said plurality of quantizers.
6 A moving picture decoding apparatus according to claim 1, wherein said compression means and expansion means conduct compression and expansion, respectively, in accordance with an orthogonal translation encoding system.
7 A moving picture decoding apparatus according to claim 1, wherein said access width control means conducts control using information included in the compressed stream.
8 A moving picture decoding apparatus according to claim 1, wherein said storage means is a frame memory.
9 A moving picture decoding apparatus to which a compressed stream generated using an inter-frame prediction system is input, said apparatus comprises:
compression means for compressing a decoded image;
storage means for storing a compressed image in said compression means;
expansion means for expanding the compressed image stored in said storage means;
quantization control means for controlling quantization on compression in said compression means; and an access width control means for applying bit allocation control to said quantization control means so as to be in conformity with the bit number of an access unit of said storage means, and said quantization control means controls quantization in said compression means based on access width information from said access width control means so that generated information content for one or a plurality of compression processing units or for every control unit of compression processing is equal to or less than the bit number of an access unit of said storage means in case that the generated information content exceeds the bit number of an access unit of said storage means or is lacking.
A moving picture decoding apparatus according to claim 9, wherein said access width control means applies bit allocation control to said quantization control means so as to be in conformity with the bit number of an access unit of said storage means, based on an occupied content of said storage means.
11 A moving picture decoding apparatus according to claim 9, wherein said access width control means conducts control using information included in the compressed stream.
12 A moving picture decoding apparatus according to claim 9, wherein said access width control means applies control to said quantization control means so that, compared with the bit number of an access unit of said storage means, in case that the allocation bit number of a coded data of a compression processing unit exceeds the bit number of an access unit of said storage means or less than the bit number of an access unit of said storage means, the allocation bit number is equal to or less than the bit number of an access unit of said storage means by subtracting a predetermined number of bits from the allocation bit number of a coded data of said compression processing unit or by increasing the allocation bit number by the predetermined number of bits, whereby the coded data can be taken out by means of one access to said storage means.
13 A moving picture decoding apparatus according to claim 9, wherein said compression means controls quantization characteristic of quantizer for quantizing said decoded image, based on control of said quantization control means.
14 A moving picture decoding apparatus according to claim 9, further comprises a plurality of quantizers having quantization characteristics different from each other, and a quantization characteristic table is shared by said plurality of quantizers.
15 A moving picture decoding apparatus according to claim 9, wherein said compression means comprises a subtracter, a quantizer, an encoder, an inverse quantizer, an adder, and predictor, a prediction error that is obtained in said subtracter by a subtraction between said decoded image and a predicted value from said predictor is supplied to said quantizer, under control of said quantization control means, said quantizer quantizes said prediction error and supplies it to said encoder and said inverse quantizer, said encoder encodes an output from said quantizer and outputs it to said storage means, and inverse quantization and local decoding are conducted in said inverse quantizer, said adder, and said predictor.
16 A moving picture decoding apparatus according to claim 9, wherein said storage means is a frame memory.
17 A moving picture decoding method comprising the steps of:
detecting the coded bit number for one or a plurality of compression processing units or for every control unit of compression processing and controlling said coded bit number so that said coded bit number is in conformity with the bit number of an access unit of a storage means when said detected coded bit number exceeds the bit number of an access unit of said storage means or is lacking.
18 A moving picture decoding method according to claim 17, wherein said step of controlling comprises a step of controlling using information included in the compressed stream.
compression means for compressing a decoded image;
storage means for storing a compressed image in said compression means;
expansion means for expanding the compressed image stored in said storage means;
quantization control means for controlling quantization on compression in said compression means; and an access width control means for applying bit allocation control to said quantization control means so as to be in conformity with the bit number of an access unit of said storage means, and said quantization control means controls quantization in said compression means based on access width information from said access width control means so that generated information content for one or a plurality of compression processing units or for every control unit of compression processing is equal to or less than the bit number of an access unit of said storage means in case that the generated information content exceeds the bit number of an access unit of said storage means or is lacking.
A moving picture decoding apparatus according to claim 9, wherein said access width control means applies bit allocation control to said quantization control means so as to be in conformity with the bit number of an access unit of said storage means, based on an occupied content of said storage means.
11 A moving picture decoding apparatus according to claim 9, wherein said access width control means conducts control using information included in the compressed stream.
12 A moving picture decoding apparatus according to claim 9, wherein said access width control means applies control to said quantization control means so that, compared with the bit number of an access unit of said storage means, in case that the allocation bit number of a coded data of a compression processing unit exceeds the bit number of an access unit of said storage means or less than the bit number of an access unit of said storage means, the allocation bit number is equal to or less than the bit number of an access unit of said storage means by subtracting a predetermined number of bits from the allocation bit number of a coded data of said compression processing unit or by increasing the allocation bit number by the predetermined number of bits, whereby the coded data can be taken out by means of one access to said storage means.
13 A moving picture decoding apparatus according to claim 9, wherein said compression means controls quantization characteristic of quantizer for quantizing said decoded image, based on control of said quantization control means.
14 A moving picture decoding apparatus according to claim 9, further comprises a plurality of quantizers having quantization characteristics different from each other, and a quantization characteristic table is shared by said plurality of quantizers.
15 A moving picture decoding apparatus according to claim 9, wherein said compression means comprises a subtracter, a quantizer, an encoder, an inverse quantizer, an adder, and predictor, a prediction error that is obtained in said subtracter by a subtraction between said decoded image and a predicted value from said predictor is supplied to said quantizer, under control of said quantization control means, said quantizer quantizes said prediction error and supplies it to said encoder and said inverse quantizer, said encoder encodes an output from said quantizer and outputs it to said storage means, and inverse quantization and local decoding are conducted in said inverse quantizer, said adder, and said predictor.
16 A moving picture decoding apparatus according to claim 9, wherein said storage means is a frame memory.
17 A moving picture decoding method comprising the steps of:
detecting the coded bit number for one or a plurality of compression processing units or for every control unit of compression processing and controlling said coded bit number so that said coded bit number is in conformity with the bit number of an access unit of a storage means when said detected coded bit number exceeds the bit number of an access unit of said storage means or is lacking.
18 A moving picture decoding method according to claim 17, wherein said step of controlling comprises a step of controlling using information included in the compressed stream.
Applications Claiming Priority (2)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| JP10-173179 | 1998-06-19 | ||
| JP17317998A JP3271585B2 (en) | 1998-06-19 | 1998-06-19 | Video decoding device |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CA2275563A1 CA2275563A1 (en) | 1999-12-19 |
| CA2275563C true CA2275563C (en) | 2003-08-05 |
Family
ID=15955563
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CA 2275563 Expired - Fee Related CA2275563C (en) | 1998-06-19 | 1999-06-18 | Moving picture decoding apparatus and moving picture decoding method |
Country Status (4)
| Country | Link |
|---|---|
| US (1) | US7333542B1 (en) |
| EP (1) | EP0975178A3 (en) |
| JP (1) | JP3271585B2 (en) |
| CA (1) | CA2275563C (en) |
Families Citing this family (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US7176152B2 (en) | 2004-06-09 | 2007-02-13 | Ferro Corporation | Lead-free and cadmium-free conductive copper thick film pastes |
| WO2007091588A1 (en) * | 2006-02-09 | 2007-08-16 | Nec Corporation | Dynamic image decoding device, decoded image recording device, their method, and program |
| JP4734168B2 (en) * | 2006-05-09 | 2011-07-27 | 株式会社東芝 | Image decoding apparatus and image decoding method |
| KR101365444B1 (en) * | 2007-11-19 | 2014-02-21 | 삼성전자주식회사 | Method and apparatus for encoding/decoding moving image efficiently through adjusting a resolution of image |
| KR101446771B1 (en) | 2008-01-30 | 2014-10-06 | 삼성전자주식회사 | Apparatus of encoding image and apparatus of decoding image |
| KR20110060181A (en) * | 2009-11-30 | 2011-06-08 | 한국전자통신연구원 | Apparatus and method for lossless/near-lossless image compression |
| JP7359653B2 (en) * | 2019-11-06 | 2023-10-11 | ルネサスエレクトロニクス株式会社 | Video encoding device |
Family Cites Families (15)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| DE4416967A1 (en) * | 1994-05-13 | 1995-11-16 | Thomson Brandt Gmbh | Method and device for transcoding bit streams with video data |
| EP0687112B1 (en) * | 1994-06-08 | 2006-09-20 | Matsushita Electric Industrial Co., Ltd. | Image conversion apparatus |
| JP3050047B2 (en) * | 1994-06-24 | 2000-06-05 | 日本ビクター株式会社 | Moving picture coding apparatus and moving picture coding method |
| EP0778709B1 (en) * | 1995-12-04 | 2001-05-23 | STMicroelectronics S.r.l. | MPEG-2 decoding with a reduced RAM requisite by ADPCM recompression before storing MPEG decompressed data |
| US5847762A (en) * | 1995-12-27 | 1998-12-08 | Thomson Consumer Electronics, Inc. | MPEG system which decompresses and then recompresses MPEG video data before storing said recompressed MPEG video data into memory |
| CA2185753C (en) | 1996-03-04 | 2000-09-12 | Hideo Ohira | Digital image decoding apparatus |
| JP3575508B2 (en) * | 1996-03-04 | 2004-10-13 | Ｋｄｄｉ株式会社 | Encoded video playback device |
| US6208689B1 (en) * | 1996-03-04 | 2001-03-27 | Mitsubishi Denki Kabushiki Kaisha | Method and apparatus for digital image decoding |
| JPH09247671A (en) | 1996-03-04 | 1997-09-19 | Mitsubishi Electric Corp | Digital image decoder |
| JP3870439B2 (en) | 1996-03-22 | 2007-01-17 | 三菱電機株式会社 | Motion compensated prediction decoding apparatus |
| US5818530A (en) | 1996-06-19 | 1998-10-06 | Thomson Consumer Electronics, Inc. | MPEG compatible decoder including a dual stage data reduction network |
| EP0817498B1 (en) | 1996-06-28 | 2001-08-22 | STMicroelectronics S.r.l. | MPEG-2 decoding with a reduced RAM requisite by ADPCM recompression before storing MPEG-2 decompressed data optionally after a subsampling algorithm |
| JPH10243402A (en) * | 1997-02-27 | 1998-09-11 | Toshiba Corp | Image processor and image processing method |
| US6205250B1 (en) * | 1998-08-27 | 2001-03-20 | Thomson Licensing S.A. | System and method for minimizing clock cycles lost to overhead data in a video decoder |
| JP2001160967A (en) * | 1999-12-03 | 2001-06-12 | Nec Corp | Image-coding system converter and coding rate converter |
-
1998
- 1998-06-19 JP JP17317998A patent/JP3271585B2/en not_active Expired - Fee Related
-
1999
- 1999-06-16 US US09/334,354 patent/US7333542B1/en not_active Expired - Lifetime
- 1999-06-18 CA CA 2275563 patent/CA2275563C/en not_active Expired - Fee Related
- 1999-06-18 EP EP19990250196 patent/EP0975178A3/en not_active Withdrawn
Also Published As
| Publication number | Publication date |
|---|---|
| EP0975178A3 (en) | 2003-08-27 |
| JP2000013801A (en) | 2000-01-14 |
| US7333542B1 (en) | 2008-02-19 |
| EP0975178A2 (en) | 2000-01-26 |
| JP3271585B2 (en) | 2002-04-02 |
| CA2275563A1 (en) | 1999-12-19 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| EP0677969B1 (en) | Moving picture coding method and an apparatus therefor | |
| JP3856262B2 (en) | Motion compensation encoding apparatus, motion compensation encoding method, and motion compensation code recording medium | |
| JP3380810B2 (en) | Image coding method | |
| US7881373B2 (en) | Video data compression apparatus and method of same | |
| US7925108B2 (en) | Encoding device and dynamic image recording system having the encoding device | |
| JP2963416B2 (en) | Video encoding method and apparatus for controlling bit generation amount using quantization activity | |
| EP0399487A2 (en) | Transformation coding device | |
| KR20020026198A (en) | Video compression | |
| JPH09191448A (en) | Video signal coder and its method | |
| CA2275563C (en) | Moving picture decoding apparatus and moving picture decoding method | |
| JP2824222B2 (en) | Video data compensation method and compensation device | |
| JP3069144B2 (en) | Video encoding device | |
| US6845179B1 (en) | Motion predicted image signal compression | |
| JP3937248B2 (en) | Video signal encoding method and video signal encoding apparatus | |
| JPH10108197A (en) | Image coder, image coding control method, and medium storing image coding control program | |
| JPH07131789A (en) | Picture coding system | |
| KR0123092B1 (en) | Method and apparatus for coding a code indicating picture starting location | |
| KR100234311B1 (en) | Video data encoding method and circuit thereof | |
| EP0927954A2 (en) | Image signal compression coding method and apparatus | |
| KR100228542B1 (en) | Adaptive bit allocation method for controlling bit generation in video encoder | |
| KR970003797B1 (en) | Encoding method for video bit stream | |
| Fogg et al. | ISO/IEC software implementation of MPEG-1 video | |
| JPH07327200A (en) | Decoder | |
| JPH05300379A (en) | Code quantity distribution circuit | |
| JPH05130580A (en) | Variable length encoding and decoding device |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| EEER | Examination request | ||
| MKLA | Lapsed |