WO2011142277A1 - 予測画像生成装置、動画像復号装置、及び、動画像符号化装置 - Google Patents
予測画像生成装置、動画像復号装置、及び、動画像符号化装置 Download PDFInfo
- Publication number
- WO2011142277A1 WO2011142277A1 PCT/JP2011/060428 JP2011060428W WO2011142277A1 WO 2011142277 A1 WO2011142277 A1 WO 2011142277A1 JP 2011060428 W JP2011060428 W JP 2011060428W WO 2011142277 A1 WO2011142277 A1 WO 2011142277A1
- Authority
- WO
- WIPO (PCT)
- Prior art keywords
- image
- decoded
- prediction
- image generation
- weighting factor
- Prior art date
Links
- 239000013598 vector Substances 0.000 claims description 161
- 238000000034 method Methods 0.000 claims description 127
- 238000013139 quantization Methods 0.000 claims description 125
- 238000005192 partition Methods 0.000 claims description 101
- 230000008569 process Effects 0.000 claims description 55
- 238000004364 calculation method Methods 0.000 abstract description 198
- 238000005303 weighing Methods 0.000 abstract 1
- 238000012545 processing Methods 0.000 description 44
- 230000015654 memory Effects 0.000 description 39
- 230000014509 gene expression Effects 0.000 description 35
- 238000010586 diagram Methods 0.000 description 28
- 238000001514 detection method Methods 0.000 description 14
- 230000008859 change Effects 0.000 description 13
- 230000006870 function Effects 0.000 description 9
- VZSRBBMJRBPUNF-UHFFFAOYSA-N 2-(2,3-dihydro-1H-inden-2-ylamino)-N-[3-oxo-3-(2,4,6,7-tetrahydrotriazolo[4,5-c]pyridin-5-yl)propyl]pyrimidine-5-carboxamide Chemical compound C1C(CC2=CC=CC=C12)NC1=NC=C(C=N1)C(=O)NCCC(N1CC2=C(CC1)NN=N2)=O VZSRBBMJRBPUNF-UHFFFAOYSA-N 0.000 description 8
- 230000000694 effects Effects 0.000 description 8
- 101100029969 Plasmodium berghei (strain Anka) PMVI gene Proteins 0.000 description 7
- 238000004891 communication Methods 0.000 description 7
- 230000007423 decrease Effects 0.000 description 7
- 230000009467 reduction Effects 0.000 description 7
- 230000002457 bidirectional effect Effects 0.000 description 6
- 238000006243 chemical reaction Methods 0.000 description 6
- 238000012217 deletion Methods 0.000 description 6
- 230000037430 deletion Effects 0.000 description 6
- 238000004519 manufacturing process Methods 0.000 description 6
- HMUNWXXNJPVALC-UHFFFAOYSA-N 1-[4-[2-(2,3-dihydro-1H-inden-2-ylamino)pyrimidin-5-yl]piperazin-1-yl]-2-(2,4,6,7-tetrahydrotriazolo[4,5-c]pyridin-5-yl)ethanone Chemical compound C1C(CC2=CC=CC=C12)NC1=NC=C(C=N1)N1CCN(CC1)C(CN1CC2=C(CC1)NN=N2)=O HMUNWXXNJPVALC-UHFFFAOYSA-N 0.000 description 5
- LDXJRKWFNNFDSA-UHFFFAOYSA-N 2-(2,4,6,7-tetrahydrotriazolo[4,5-c]pyridin-5-yl)-1-[4-[2-[[3-(trifluoromethoxy)phenyl]methylamino]pyrimidin-5-yl]piperazin-1-yl]ethanone Chemical compound C1CN(CC2=NNN=C21)CC(=O)N3CCN(CC3)C4=CN=C(N=C4)NCC5=CC(=CC=C5)OC(F)(F)F LDXJRKWFNNFDSA-UHFFFAOYSA-N 0.000 description 5
- YLZOPXRUQYQQID-UHFFFAOYSA-N 3-(2,4,6,7-tetrahydrotriazolo[4,5-c]pyridin-5-yl)-1-[4-[2-[[3-(trifluoromethoxy)phenyl]methylamino]pyrimidin-5-yl]piperazin-1-yl]propan-1-one Chemical compound N1N=NC=2CN(CCC=21)CCC(=O)N1CCN(CC1)C=1C=NC(=NC=1)NCC1=CC(=CC=C1)OC(F)(F)F YLZOPXRUQYQQID-UHFFFAOYSA-N 0.000 description 5
- 238000013459 approach Methods 0.000 description 5
- 230000002123 temporal effect Effects 0.000 description 4
- MKYBYDHXWVHEJW-UHFFFAOYSA-N N-[1-oxo-1-(2,4,6,7-tetrahydrotriazolo[4,5-c]pyridin-5-yl)propan-2-yl]-2-[[3-(trifluoromethoxy)phenyl]methylamino]pyrimidine-5-carboxamide Chemical compound O=C(C(C)NC(=O)C=1C=NC(=NC=1)NCC1=CC(=CC=C1)OC(F)(F)F)N1CC2=C(CC1)NN=N2 MKYBYDHXWVHEJW-UHFFFAOYSA-N 0.000 description 3
- NIPNSKYNPDTRPC-UHFFFAOYSA-N N-[2-oxo-2-(2,4,6,7-tetrahydrotriazolo[4,5-c]pyridin-5-yl)ethyl]-2-[[3-(trifluoromethoxy)phenyl]methylamino]pyrimidine-5-carboxamide Chemical compound O=C(CNC(=O)C=1C=NC(=NC=1)NCC1=CC(=CC=C1)OC(F)(F)F)N1CC2=C(CC1)NN=N2 NIPNSKYNPDTRPC-UHFFFAOYSA-N 0.000 description 3
- AFCARXCZXQIEQB-UHFFFAOYSA-N N-[3-oxo-3-(2,4,6,7-tetrahydrotriazolo[4,5-c]pyridin-5-yl)propyl]-2-[[3-(trifluoromethoxy)phenyl]methylamino]pyrimidine-5-carboxamide Chemical compound O=C(CCNC(=O)C=1C=NC(=NC=1)NCC1=CC(=CC=C1)OC(F)(F)F)N1CC2=C(CC1)NN=N2 AFCARXCZXQIEQB-UHFFFAOYSA-N 0.000 description 3
- 230000006872 improvement Effects 0.000 description 3
- 230000009466 transformation Effects 0.000 description 3
- FFBHFFJDDLITSX-UHFFFAOYSA-N benzyl N-[2-hydroxy-4-(3-oxomorpholin-4-yl)phenyl]carbamate Chemical compound OC1=C(NC(=O)OCC2=CC=CC=C2)C=CC(=C1)N1CCOCC1=O FFBHFFJDDLITSX-UHFFFAOYSA-N 0.000 description 2
- 238000011156 evaluation Methods 0.000 description 2
- 230000003287 optical effect Effects 0.000 description 2
- 101100025317 Candida albicans (strain SC5314 / ATCC MYA-2876) MVD gene Proteins 0.000 description 1
- 241000557626 Corvus corax Species 0.000 description 1
- 101150079299 MVD1 gene Proteins 0.000 description 1
- 101150020780 MVD2 gene Proteins 0.000 description 1
- 230000005540 biological transmission Effects 0.000 description 1
- 238000012888 cubic function Methods 0.000 description 1
- 238000011161 development Methods 0.000 description 1
- 238000013213 extrapolation Methods 0.000 description 1
- 238000005562 fading Methods 0.000 description 1
- 230000007246 mechanism Effects 0.000 description 1
- 238000010295 mobile communication Methods 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 239000004065 semiconductor Substances 0.000 description 1
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
- H04N19/10—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
- H04N19/103—Selection of coding mode or of prediction mode
- H04N19/105—Selection of the reference unit for prediction within a chosen coding or prediction mode, e.g. adaptive choice of position and number of pixels used for prediction
-
- 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
- H04N19/134—Methods or arrangements for coding, decoding, compressing or decompressing digital video signals using adaptive coding characterised by the element, parameter or criterion affecting or controlling the adaptive coding
- H04N19/136—Incoming video signal characteristics or properties
- H04N19/137—Motion inside a coding unit, e.g. average field, frame or block difference
- H04N19/139—Analysis of motion vectors, e.g. their magnitude, direction, variance or reliability
-
- 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
- H04N19/169—Methods or arrangements for coding, decoding, compressing or decompressing digital video signals using adaptive coding characterised by the coding unit, i.e. the structural portion or semantic portion of the video signal being the object or the subject of the adaptive coding
- H04N19/17—Methods or arrangements for coding, decoding, compressing or decompressing digital video signals using adaptive coding characterised by the coding unit, i.e. the structural portion or semantic portion of the video signal being the object or the subject of the adaptive coding the unit being an image region, e.g. an object
- H04N19/176—Methods or arrangements for coding, decoding, compressing or decompressing digital video signals using adaptive coding characterised by the coding unit, i.e. the structural portion or semantic portion of the video signal being the object or the subject of the adaptive coding the unit being an image region, e.g. an object the region being a block, e.g. a macroblock
-
- 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
- H04N19/50—Methods or arrangements for coding, decoding, compressing or decompressing digital video signals using predictive coding
- H04N19/503—Methods or arrangements for coding, decoding, compressing or decompressing digital video signals using predictive coding involving temporal prediction
- H04N19/51—Motion estimation or motion compensation
- H04N19/513—Processing of motion vectors
Definitions
- the present invention relates to predicted image generation for generating a predicted image by adding an offset value to a weighted average of motion compensated images generated from each decoded image.
- the present invention also relates to a moving picture decoding apparatus and a moving picture encoding apparatus provided with such a predicted image generation apparatus.
- a moving image encoding device that encodes moving images and a moving image decoding device that decodes moving images are widely used.
- a typical moving image encoding method for example, H.264 is used.
- KTA software which is a codec for joint development in AVC (Non-Patent Document 1) and VCEG (Video Coding Expert Group).
- each picture constituting a moving image is composed of a slice obtained by dividing a picture, a macroblock obtained by dividing the slice, and a block obtained by dividing the macroblock.
- a slice obtained by dividing a picture is composed of a macroblock obtained by dividing the slice, and a block obtained by dividing the macroblock.
- a block obtained by dividing the macroblock is managed by a hierarchical structure.
- the moving image encoding device encodes a moving image in units of blocks. Specifically, (1) the prediction residual obtained by subtracting the prediction image from the original image is subjected to DCT transform (Discrete Cosine Transform) for each block, (2) the obtained DCT coefficient is quantized, and (3) The quantized DCT coefficient is variable length encoded.
- the amount of code can usually be reduced by DCT transform / quantization / encoding the prediction residual instead of DCT transform / quantization / encoding the original image.
- the moving image encoding device generates the above-described predicted image based on a locally decoded image obtained by inverse quantization / inverse DCT transform of the quantized DCT coefficient.
- the prediction image generation methods include inter-screen prediction (inter prediction) and intra-screen prediction (intra prediction), and the video encoding device selects which prediction method to use for each macroblock.
- a predicted image in each partition is generated by extrapolation from a locally decoded image in another encoded / decoded partition (hereinafter referred to as “reference partition”) belonging to the same picture as that partition.
- reference partition a predicted image in each partition is generated by motion compensated prediction from another encoded / decoded picture (hereinafter referred to as “reference image”) different from the picture to which the partition belongs.
- reference image refers to a prediction unit obtained by dividing a macroblock.
- H.C. H.264 / MPEG-4 In inter prediction in AVC, a weighted average w 1 ⁇ mc of motion compensated images mc 1 and mc 2 generated from each of reference images (local decoded images) P 1 and P 2 as predicted images to be subtracted from the encoding target image. Weighted prediction using 1 + w 2 ⁇ mc 2 plus an offset value can be performed.
- the weighted prediction includes an Explicit mode (implicit mode) and an Implicit mode (implicit mode), and the moving image encoding apparatus can select which mode to use for each partition.
- the weight coefficient w 1 and w 2 are encoded in the moving image encoding device, and the weight coefficient w 1 and w 2 are decoded in the moving image decoding device.
- the prediction target image (the encoding target image in the video encoding device and the decoding target image in the video decoding device) P is referred to in both the video encoding device and the video decoding device.
- the weight coefficients w 1 and w 2 are set according to the interframe distance between images P 1 and P 2 (locally decoded images in the moving image encoding device and decoded images in the moving image decoding device).
- the inter-frame distance between the reference image P 1 and the reference image P 2 td, reference image P 1 and the distance between the prediction target image P as tb, the motion generated from the reference image P 2 compensation image mc 2 set the value of the weighting factor w 2 to tb / td, sets the value of the weighting factor w 1 for multiplying the motion compensated image mc 1 generated from the reference image P 1 to 1-w 2 to be multiplied by the.
- a and B are code amounts that are standardized so that the SN ratio (PSNR) matches between the case where weighted prediction is not performed and the case where weighted prediction in Implicit mode is performed.
- the weighting factor of weighted prediction in Implicit mode is a weighting factor that is optimal (the prediction accuracy is highest) when compensating for a steady temporal change caused by fading or dissolve. Therefore, when the temporal change is steady, the decrease in the code amount due to the improvement of the prediction accuracy surpasses the increase in the code amount due to the remaining coding noise, and the code amount decreases as a whole. On the other hand, when the temporal change is not steady, a decrease in the code amount due to improvement in prediction accuracy may be less than an increase in the code amount due to residual coding noise, and the code amount may increase as a whole. It is considered that the increase in the code amount in the three types of sequences described above is caused by such a mechanism.
- the present invention has been made in view of the above problems, and an object of the present invention is to provide a predicted image generation apparatus and a prediction image that generate a predicted image by adding an offset value to a weighted average of motion compensated images generated from each of the reference images.
- the image generation method the possibility of increasing the amount of code for a sequence whose temporal change is not steady is reduced, and the more stable encoding efficiency is improved.
- the predicted image generation apparatus uses the decoded images P 1 and P 2 as reference images, motion compensation image generation means for generating motion compensated images mc 1 and mc 2 for the decoded image P, and the motion compensated image a weighting factor setting means for setting the value of the weight coefficients w 1 and w 2 to be multiplied to each of the motion generated compensated image mc 1 and mc 2 by generating means, the weighting coefficients w 1 and set by the weighting coefficient setting means Using w 2 , an offset value is added to the weighted average w 1 ⁇ mc 1 + w 2 ⁇ mc 2 of the motion compensated images mc 1 and mc 2 generated by the motion compensated image generating means, and the decoded image P
- a predicted image generation apparatus comprising: predicted image generation means for generating a predicted image Pred.
- the predicted image generation method includes a motion compensated image generation step for generating motion compensated images mc 1 and mc 2 for the decoded image P using each of the decoded images P 1 and P 2 as a reference image, and the motion described above.
- a weighting factor setting step of setting the value of the weight coefficients w 1 and w 2 to be multiplied to each of the motion generated by compensated image generation step compensated image mc 1 and mc 2, set by the weighting factor setting step weights By adding the offset value to the weighted average w 1 ⁇ mc 1 + w 2 ⁇ mc 2 of the motion compensation images mc 1 and mc 2 generated in the motion compensation image generation step using the coefficients w 1 and w 2 .
- a predicted image generation step of generating a predicted image Pred for the decoded image P.
- the predicted image generation apparatus uses the weight coefficient setting means to set the inter-frame distance between the decoded image P 1 and the decoded image P 2 to be td, and the decoded image P 1 and the decoded image.
- the distance from P is tb, and the value of the weighting factor w 2 is set to satisfy 0 ⁇
- the predicted image generation method sets the interframe distance between the decoded image P 1 and the decoded image P 2 to be td and the decoded image P 1 and the decoded image in the weight coefficient setting step.
- the distance from the image P is tb, and the weight coefficient w 2 is set so that 0 ⁇
- the weighting factor w 2 is set so that 0 ⁇
- the weighting coefficient w 2 finally multiplied by the motion compensated image mc 2 satisfies the condition 0 ⁇
- the pixel value of the predicted image is obtained by dividing the pixel value obtained by adding the offset value to the weighted average W 1 ⁇ mc 1 + W 2 ⁇ mc 2 using the integer coefficients W 1 and W 2 by the constant C.
- the weight coefficient w 2 W 2 / C only needs to satisfy the condition 0 ⁇
- the offset value may be set to 0. That is, a configuration in which the offset value is not substantially added is also included in the scope of the present invention.
- the present invention has an effect of reducing the possibility that the amount of code increases due to prediction when encoding a prediction residual obtained by subtracting a prediction image from an encoding target image.
- FIG. 2 is a diagram illustrating a configuration of encoded data decoded by the video decoding device in FIG. 1, wherein (a) is a diagram illustrating a configuration of a picture layer of the encoded data, and (b) is a diagram illustrating a picture layer; It is a figure which shows the structure of the slice layer contained, (c) is a figure which shows the structure of the macroblock layer contained in a slice layer, (d) is a partition layer contained in a macroblock layer, It is a figure which shows the structure of the partition layer about the partition where inter prediction is used, (e) is a partition layer contained in a macroblock layer, Comprising: The figure which shows the structure of the partition layer about the partition where intra prediction is used It is.
- Embodiment 1 (Configuration of encoded data) Prior to the description of the video encoding device 2 and the video decoding device 1 according to the present embodiment, the data structure of the encoded data generated by the video encoding device 2 and decoded by the video decoding device 1 will be described. I do.
- the encoded data generated by the moving image encoding device 2 and decoded by the moving image decoding device 1 is a hierarchy including a sequence layer, a GOP (Group Of Pictures) layer, a picture layer, a slice layer, and a macroblock layer. It has a structure.
- Fig. 2 shows the structure of the encoded data for the layers below the picture layer.
- 2A to 2F show the structures of the picture layer P, the slice layer S, the macroblock layer MB, the macroblock header MBH, the prediction parameter PP of the inter prediction partition, and the prediction parameter PP of the intra prediction partition, respectively.
- FIG. 1 shows the structure of the encoded data for the layers below the picture layer.
- 2A to 2F show the structures of the picture layer P, the slice layer S, the macroblock layer MB, the macroblock header MBH, the prediction parameter PP of the inter prediction partition, and the prediction parameter PP of the intra prediction partition, respectively.
- the picture layer P is a set of data referred to by the video decoding device 1 in order to decode the corresponding picture. As shown in FIG. 2A, the picture layer P includes a picture header PH and slice layers S 1 to S Ns (Ns is the total number of slice layers included in the picture layer P).
- the picture header PH includes a coding parameter group referred to by the video decoding device 1 in order to determine the decoding method of the corresponding picture.
- the encoding mode information (entoropy_coding_mode_flag) indicating the variable length encoding mode used in encoding by the moving image encoding device 2 is an example of an encoding parameter included in the picture header PH.
- Each slice layer S included in the picture layer P is a set of data referred to by the video decoding device 1 in order to decode the corresponding slice.
- the slice layer S includes a slice header SH and macroblock layers MB 1 to MB Nm (Nm is the total number of macroblocks included in the slice S).
- the slice header SH includes a coding parameter group that the moving image decoding apparatus 1 refers to in order to determine a decoding method for the corresponding slice.
- Slice type designation information (slice_type) for designating a slice type
- POC designation information (pic_order_cnt_lbs, delta_pic_order_cnt, etc.) for designating the display order (POC: Picture Order Count) of a picture including the corresponding slice
- POC designation information (pic_order_cnt_lbs, delta_pic_order_cnt, etc.) for designating the display order (POC: Picture Order Count) of a picture including the corresponding slice
- POC designation information (pic_order_cnt_lbs, delta_pic_order_cnt, etc.) for designating the display order (POC: Picture Order Count) of a picture including the corresponding slice
- the weighting factor designation information (pred_weight_table) that designates the weighting factor used when the moving image coding apparatus 2 performs the
- slice types that can be specified by the slice type specification information, (1) I slice using only intra prediction at the time of encoding, and (2) P using unidirectional prediction or intra prediction at the time of encoding. Slice, (3) B-slice using unidirectional prediction, bidirectional prediction, or intra prediction at the time of encoding.
- Each macroblock layer MB included in the slice layer S is a set of data that the video decoding device 1 refers to in order to decode the corresponding macroblock.
- the macroblock layer MB includes a skip flag SKIP, a macroblock header MBH, prediction parameters PP 1 to PP NP , and quantization prediction errors QD 1 to QD Nb .
- Np represents the total number of partitions (prediction units) included in the corresponding macroblock
- Nb represents the total number of blocks (conversion units) included in the corresponding macroblock.
- the skip flag SKIP When the value of the skip flag SKIP is 1, that is, when the corresponding macroblock is a skip block, the macroblock header MBH, the prediction parameters PP 1 to PP NP , and the quantized prediction errors QD 1 to QD in that macroblock layer Nb is omitted.
- the macroblock header MBH includes an encoding parameter referred to by the video decoding device 1 in order to determine a decoding method of the corresponding macroblock.
- macroblock type designation information MBT (mb_type) for designating the macroblock type of the corresponding macroblock
- CBP coded_block_pattern
- quantization step Quantization parameter difference ⁇ qp (mb_qp_delta) that specifies the size of.
- the macroblock type designation information MBT includes prediction unit designation information PT and conversion unit designation information TT.
- the prediction unit designation information PT includes the division pattern of the corresponding macroblock into partitions (prediction units) and the prediction method (L0 unidirectional prediction, L1) used when the moving image encoding apparatus 2 generates a prediction image in each partition. Unidirectional prediction, bidirectional prediction, etc.).
- the partition is a square region of 16 ⁇ 16 pixels, 8 ⁇ 8 pixels, or 4 ⁇ 4 pixels, or a rectangular region of 16 ⁇ 8 pixels, 8 ⁇ 16 pixels, 8 ⁇ 4 pixels, or 4 ⁇ 8 pixels.
- the conversion unit designation information TT designates the division pattern of the corresponding macroblock into blocks (conversion units).
- the block is a square area of 16 ⁇ 16 pixels, 8 ⁇ 8 pixels, or 4 ⁇ 4 pixels.
- the partition when the macroblock block is a square area of 16 ⁇ 16 pixels is illustrated, but the present invention is not limited to this.
- a 64 ⁇ 64 pixel macroblock a 64 ⁇ 64 pixel or 32 ⁇ 32 pixel square area, or 64 ⁇ 32 pixel, 32 ⁇ 64 pixel, 32 ⁇ 16 pixel, or 16 ⁇ 32 pixel
- the rectangular area is also allowed as a partition.
- the quantization parameter difference ⁇ qp is a difference qp ⁇ qp ′ between the quantization parameter qp in the corresponding macroblock and the quantization parameter qp ′ in the macroblock encoded immediately before the macroblock.
- Each quantized prediction residual QD n included in the macroblock layer MB is encoded data generated by the moving image encoding apparatus 2 performing the following processes 1 to 3 on the corresponding block.
- Process 1 DCT transform (Discrete Cosine Transform) is performed on the prediction residual obtained by subtracting the prediction image from the encoding target image.
- Process 2 The DCT coefficient obtained in Process 1 is quantized.
- Process 3 The DCT coefficient quantized in Process 2 is variable length encoded.
- the prediction parameter PP related to the inter prediction partition in which the prediction image is generated by the inter prediction includes a reference image index RI, an estimated motion vector, as shown in FIG.
- the index PMVI and the motion vector residual MVD are included.
- the motion vector residual MVD is encoded data generated by the moving image encoding device 2 executing the following processes 4 to 6.
- Process 4 Select an encoded / decoded local decoded image, and derive a motion vector mv for the corresponding partition with reference to the selected encoded / decoded local decoded image (hereinafter also referred to as “reference image”). To do.
- Process 5 An estimation method is selected, and an estimated value (hereinafter also referred to as “estimated motion vector”) pmv of a motion vector mv assigned to the corresponding partition is derived using the selected estimation method.
- Process 6 The motion vector residual MVD obtained by subtracting the estimated motion vector pmv derived in Process 5 from the motion vector mv derived in Process 4 is encoded.
- the reference image index RI described above specifies the locally decoded image (reference image) that has been encoded / decoded selected in the process 4, and the estimated motion vector index PMVI described above is selected in the process 5. Specifies the estimation method.
- estimation methods that can be selected in the process 5 (1) in a locally decoded image being encoded / decoded, motion vectors assigned to partitions adjacent to corresponding partitions (hereinafter also referred to as “adjacent partitions”) A method in which the median is an estimated motion vector pmv, or (2) motion allocated to a partition (often referred to as a “collocated partition”) occupying the same position as the corresponding partition in an encoded / decoded local decoded image
- a method using the vector as the estimated motion vector pmv can be used.
- the prediction parameter PP related to the partition for which unidirectional prediction is performed includes one reference image index RI, one estimated motion vector index PMVI, and one motion vector residual MVD.
- the prediction parameters PP for a partition that performs bi-directional prediction include two reference image indexes RI1 and RI2, two estimated motion vector indexes PMVI1 and PMVI2, and two motion vector residuals MVD1. And MVD2.
- the prediction parameter PP related to an intra prediction partition in which a prediction image is generated by intra prediction includes a prediction mode flag PMF and a prediction mode code PMC, as shown in FIG. Including.
- the prediction mode flag PMF takes the value 1 when the prediction mode in the corresponding partition matches the prediction mode estimated from the adjacent partition, and takes the value 0 otherwise.
- the prediction mode code PMC is obtained by encoding prediction mode specifying information for specifying a prediction mode in the corresponding partition, and is omitted when the value of the prediction mode flag PMF is 1.
- the moving picture decoding apparatus 1 includes H.264 as a part thereof. H.264 / MPEG-4. It is a decoding device including a technique adopted in AVC and KTA software.
- FIG. 1 is a block diagram showing a configuration of the moving picture decoding apparatus 1.
- the moving picture decoding apparatus 1 includes a variable length code decoding unit 13, a motion vector restoration unit 14, a buffer memory 15, an inter prediction image generation unit 16, an intra prediction image generation unit 17, and a prediction method determination unit 18. , An inverse quantization / inverse transform unit 19 and an adder 20 are provided.
- the moving picture decoding apparatus 1 is an apparatus for generating moving picture # 2 by decoding encoded data # 1.
- the variable length code decoding unit 13 decodes the prediction parameter PP related to each partition from the encoded data # 1. That is, for the inter prediction partition, the reference image index RI, the estimated motion vector index PMVI, and the motion vector residual MVD are decoded from the encoded data # 1, and these are supplied to the motion vector restoration unit 14. On the other hand, regarding the intra prediction partition, the prediction mode designation information is decoded from the encoded data # 1, and this is supplied to the intra predicted image generation unit 17. Further, the variable length code decoding unit 13 decodes the macroblock type MBT from the encoded data, and supplies this to the prediction method determination unit 18 (not shown).
- variable length code decoding unit 13 decodes the quantization prediction residual QD for each block and the quantization parameter difference ⁇ qp for the macroblock including the block from the encoded data # 1, and dequantizes and decodes them. This is supplied to the inverse conversion unit 19.
- the motion vector restoration unit 14 restores the motion vector mv related to each inter prediction partition from the motion vector residual MVD related to that partition and the restored motion vector mv ′ related to another partition. Specifically, (1) the estimated motion vector pmv is derived from the restored motion vector mv ′ according to the estimation method specified by the estimated motion vector index PMVI, and (2) the derived estimated motion vector pmv and the remaining motion vector The motion vector mv is obtained by adding the difference MVD. It should be noted that the restored motion vector mv ′ relating to other partitions can be read from the buffer memory 15. The motion vector restoration unit 14 supplies the restored motion vector mv together with the corresponding reference image index RI to the inter predicted image generation unit 16. For the inter prediction partition that performs bi-directional prediction (weighted prediction), the restored two motion vectors mv1 and mv2 are supplied to the inter prediction image generation unit 16 together with the corresponding reference image indexes RI1 and RI2.
- the inter prediction image generation unit 16 generates a motion compensation image mc related to each inter prediction partition. Specifically, using the motion vector mv supplied from the motion vector restoration unit 14, a motion compensated image mc is generated from the decoded image P ′ designated by the reference image index RI supplied from the motion vector restoration unit 14 as well. To do.
- the decoded image P ′ is a decoded image that has already been decoded, and the inter predicted image generation unit 16 can read out the pixel value of each pixel constituting the decoded image P ′ from the buffer memory 15. it can.
- the motion compensation image mc generated by the inter prediction image generation unit 16 is supplied to the prediction method determination unit 18 as an inter prediction image Pred_Inter.
- a motion compensated image mc1 is generated from the decoded image P1 specified by the reference image index RI1 using the motion vector mv1, and (2 )
- a motion compensation image mc2 is generated from the reference image P2 specified by the reference image index RI2 using the motion vector mv2, and (3) an offset value is added to the weighted average of the motion compensation image mc1 and the motion compensation image mc2.
- the inter prediction image Pred_Inter is generated.
- a method for setting the weighting factors w1 and w2 to be multiplied by each of the two motion compensation images mc1 and mc2 will be described later in place of the referenced drawing.
- the intra predicted image generation unit 17 generates a predicted image Pred_Intra related to each intra prediction partition. Specifically, the prediction image Pred_Intra is generated from the decoded image P using the prediction mode specified by the prediction mode specification information supplied from the variable length code decoding unit 13.
- the decoded image P is a decoded image that has not been decoded yet
- the intra-predicted image generation unit 17 includes each pixel belonging to the block that has already been decoded among the pixels constituting the decoded image P.
- the pixel value of the pixel can be read from the buffer memory 15.
- the intra predicted image Pred_Intra generated by the intra predicted image generation unit 17 is supplied to the prediction method determination unit 18.
- the prediction method determination unit 18 determines whether each partition is an inter prediction partition that should perform inter prediction or an intra prediction partition that should perform intra prediction, based on the macroblock type MBT. In the former case, the inter predicted image Pred_Inter generated by the inter predicted image generation unit 16 is supplied to the adder 20 as the predicted image Pred. In the latter case, the inter predicted image generation unit 17 generates the inter predicted image Pred_Inter. The intra predicted image Pred_Intra that has been processed is supplied to the adder 20 as the predicted image Pred.
- the inverse quantization / inverse transform unit 19 (1) inversely quantizes the quantized prediction residual QD, (2) performs inverse DCT (Discrete Cosine Transform) transform on the DCT coefficient obtained by the inverse quantization, and (3) The prediction residual D obtained by the inverse DCT transform is supplied to the adder 20.
- the inverse quantization / inverse transform unit 19 derives the quantization step QP from the quantization parameter difference ⁇ qp supplied from the variable length code decoding unit 13.
- the adder 20 generates the decoded image P by adding the prediction image Pred supplied from the prediction method determination unit 18 and the prediction residual D supplied from the inverse quantization / inverse transformation unit 19.
- the decoded image P generated by the adder 20 is output to the outside and stored in the buffer memory 15 in association with the POC designation information decoded from the encoded data by the variable length code decoding unit 13.
- FIG. 3 is a block diagram illustrating a configuration of the inter predicted image generation unit 16.
- the inter prediction image generation unit 16 includes a reference value calculation unit 16a, a weight coefficient calculation unit 16b, a motion compensation unit 16c, and a prediction image generation unit 16d.
- generation part 16 can perform unidirectional prediction and bidirectional
- the reference value calculation unit 16a reads POC designation information associated with decoded images P1 and P2 (hereinafter referred to as reference images L0 and L1) designated by the reference image indexes RI1 and RI2 from the buffer memory 15 and reads the reference image L0. And the display order POC_L0 and POC_L1 of L1 are specified. Then, the interframe distances td and tb are calculated from the display order POC_L0 and POC_L1 of the reference images L0 and L1 and the display order POC_cur of the decoded image P being decoded, and these are supplied to the weighting factor calculation unit 16b.
- the interframe distance td is the interframe distance between the frame including the reference image L0 and the frame including the reference image L1
- the interframe distance tb is the frame including the reference image L0 and the frame including the decoded image P being decoded.
- the distance between frames A specific method of calculating the interframe distances td and tb will be described later.
- the weighting factor calculation unit 16b calculates weighting factors w1 and w2 from the interframe distances td and tb calculated by the reference value calculation unit 16a, and supplies these to the predicted image generation unit 16d. A method for calculating the weighting factors w1 and w2 will be described later.
- the motion compensation unit 16c reads the reference images L0 (decoded image P1) and L1 (decoded image P2) designated by the reference image indexes RI1 and RI2 from the buffer memory 15. Then, motion compensation images mc1 and mc2 are generated from each of the reference images L1 and L0 by motion compensation using each of the motion vectors mv1 and mv2, and these are supplied to the predicted image generation unit 16d. A method for generating the motion compensated images mc1 and mc2 will be described later.
- the prediction image generation unit 16d generates the inter prediction image Pred_Inter using the weight coefficients w1 and w2 calculated by the weight coefficient calculation unit 16b and the motion compensation images mc1 and mc2 generated by the motion compensation unit 16c. And output this.
- the reference value calculation unit 16a calculates the inter-frame distance tb from the following equation (1) using the display order POC_L0 of the reference image L0 and the display order POC_cur of the decoded image P being decoded.
- the reference value calculation unit 16a calculates the inter-frame distance td from the following equation (2) using the display order POC_L0 of the reference image L0 and the display order POC_L1 of the reference image L1.
- the inter-frame distance tb and the inter-frame distance td may be calculated using the following expressions (3) and (4).
- abs Represents taking the absolute value of the value in parentheses. That is, the interframe distance tb and the interframe distance td represented by the expressions (1) and (2) can be negative values, but the expressions (3) and (4) The inter-distance tb and the inter-frame distance td represented by are only positive values.
- the display order POC is used for calculating tb and td.
- other information that can calculate / estimate the distance (time difference) between frames such as display time and display time is used.
- Equivalent processing can be performed. For example, if the display time is T, POC_cur, POC_L0, and POC_L1 are T_cur (display time of the decoded image P being decoded), T_L0 (display time of the reference image L0), and T_L1 (display time of the reference image L1), respectively.
- Tb and td can be calculated.
- the fact that the same type of information can be used, not limited to POC also applies to the following configurations.
- the weight coefficient calculation unit 16b calculates the weight coefficients w1 and w2 used in the weighted prediction using the interframe distance tb and the interframe distance td calculated by the reference value calculation unit 16a. Specific examples of the calculation processing of the weighting factors w1 and w2 will be described in the following (Specific example 1-1) to (Specific example 1-5).
- the motion compensation unit 16c generates a motion compensated image mc1 from the reference image L0 by motion compensation using the motion vector mv1. Similarly, the motion compensation unit 16c generates a motion compensated image mc2 from the reference image L1 by motion compensation using the motion vector mv2.
- each pixel value mc1 (x, y) of the motion compensated image mc1 and each pixel value mc2 (x, y) of the motion compensated image mc2 are generated using Equation (5) and Equation (6). Is done.
- (x, y) represents the coordinates of the target pixel in the motion compensation image mc1 and the motion compensation image mc2
- (mv1x, mv1y) represents the x component and the y component of the motion vector mv1
- (Mv2x, mv2y) represents the x component and the y component of the motion vector mv2.
- L0 (x, y) represents the pixel value at the coordinates (x, y) of the reference image L0
- L1 (x, y) represents the pixel value at the coordinates (x, y) of the reference image L1. Yes.
- the predicted image generating unit 16d The pixel value pred (x, y) of each pixel is generated using Expression (7).
- the weighting factor calculation unit 16b performs any one of the calculation processes listed in the following (specific example 1-1) to (specific example 1-5), and each of the motion compensation images in Expression (7). Weight coefficients w1 and w2 to be multiplied by the pixel value are calculated. In any specific example, the ratio of the weighting factors w1 and w2 is closer to 1: 1 than 1-tb / td: tb / td. That is, 0 ⁇
- the weighting factor calculation unit 16b is given the values of the weighting factors w1 and w2 in advance when the ratio of the weighting factors w1 and w2 obtained by referring to the interframe distance is greatly different from 1: 1. Set to a value, otherwise set to the ratio of the distance between frames.
- the weighting factor calculation unit 16b uses the interframe distance tb and the interframe distance td calculated by the reference value calculation unit 16a according to the equations (1) and (2) or the equations (3) and (4),
- the weighting factors w1 and w2 are calculated according to the equations (8) and (9).
- clip3 (argument 1, argument2, argument3) takes argument 1 when argument 3 is smaller than argument 1 (argument 1> argument 3), and argument 3 is larger than argument 2
- the argument 2 is taken, and when the argument 3 is a value from the argument 1 to the argument 2 (argument 1 ⁇ argument 3 ⁇ argument 2), the function taking the argument 3 is represented. That is, clip3 (argument 1, argument2, argument3) represents clip processing that limits the range of values using two different threshold values (argument 1 and argument 2).
- the weighting factor w2 is th1 when tb / td is smaller than th1, and is th2 when tb / td is larger than th2, and when tb / td is greater than or equal to th1 and less than or equal to th2. , Tb / td.
- th1 and th2 represent a lower limit threshold and an upper limit threshold in clip processing, and the values of th1 and th2 are set in advance so that the encoding efficiency is maximized.
- the inter-frame distance ratio tb / td is close to 0.5, that is, the distance between the reference image L0 and the prediction target image (decoded image being decoded), and the reference image L1.
- the distance between the target image and the prediction target image is closer to 1: 1
- noise that may occur in the predicted image is further reduced.
- the prediction accuracy tends to be improved by using the ratio of the interframe distance as a weighting coefficient.
- the ratio of the interframe distance is used as the weighting factor, when the ratio of the interframe distance is far from 0.5, noise generated in the predicted image increases, and thus the coding efficiency is increased. There is a problem that decreases.
- the weighting factor calculation unit 16b performs the operation of this example, and uses the ratio of the interframe distance as the weighting factor, thereby improving the prediction accuracy and increasing the prediction accuracy as described above. By performing the clip processing, an increase in noise can be suppressed.
- the moving picture coding apparatus 2 having the same means as the weighting coefficient calculating unit 16b that performs the operation of this example is a case where the ratio of the interframe distance is used as the weighting coefficient as shown in FIG. Compared with the case where the clipping process is not performed, encoded data with high encoding efficiency can be generated.
- the moving picture decoding apparatus 1 including the weight coefficient calculation unit 16b that performs the operation of this example can appropriately decode the encoded data generated in such a manner with high encoding efficiency.
- the weighting factor may be expressed by an integer.
- the weighting factors w1 and w2 can be calculated by multiplying Equation (8) and Equation (9) by a predetermined integer, and each pixel value of the inter prediction image is determined in advance. It can be generated by dividing equation (7) by an integer.
- weighting factors w1 and w2 can be calculated by the following equations (8 ') and (9').
- w2a is a temporary variable used for calculating w2.
- Expressions (8 ′) and (9 ′) are obtained by multiplying Expression (8) and Expression (9) by 64, but the values to be multiplied by Expression (8) and Expression (9) are However, it is a value that is changed according to the bit depth.
- the weighting factor calculation unit 16b uses the interframe distance tb and the interframe distance td calculated by the reference value calculation unit 16a to calculate the weighting factor w2 using the equations (10) and (11).
- w2b ⁇ (tb / td ⁇ 0.5) ⁇ r + 0.5 ⁇ (10)
- w2 clip3 (th1, th2, w2b) (11)
- w2b is a variable temporarily used to calculate w2, and as is clear from the equation (10), the value of w2b is closer to 0.5 than the given value of tb / td. Takes a value.
- r in equation (10) is a constant that defines how close the value of w2b is to 0.5 compared to the value of tb / td.
- the weighting factor calculation unit 16b calculates the weighting factor w1 from the equation (9) using the equation (10) and the weighting factor w2 obtained by the equation (11).
- the weighting factor calculation unit 16b thus calculates w2 by performing clip processing on w2b that is closer to 0.5 than the given value of tb / td.
- Equation (10) the processing for bringing the inter-frame distance ratio tb / td closer to 0.5 is not limited to Equation (10), and other equations may be used.
- a high-order function such as a cubic function, a sigmoid function, a logistic function, or the like can be used.
- it may be derived by subtracting a table that gives a value closer to 0.5 from the ratio tb / td.
- the weighting factor w2 closer to 0.5 can be calculated for a given tb / td as compared to the case of the specific example 1-1. There is an effect that generated noise is further reduced.
- the variable w2b becomes a value in the range between the lower limit threshold and the upper limit threshold of clip processing. Since the weighting factor w2 reflecting the distance ratio more effectively can be calculated, the prediction error can be further reduced. Moreover, since the weighting factor w2 approaches 0.5 by this process, the weighting factor w1 also becomes a value close to 0.5. That is, since the ratio between the weighting factor w1 and the weighting factor w2 approaches 1: 1, it is possible to more efficiently suppress a decrease in encoding efficiency due to noise.
- the lower limit threshold and the upper limit threshold for clip processing are set according to the partition size, and the weighting factors w1 and w2 are calculated.
- the partition size refers to the partition size of a prediction target partition (partition where a prediction image should be generated).
- the weighting factor calculation unit 16b can specify the partition size of the prediction target partition by referring to the macroblock type MBT decoded by the variable length code decoding unit 13.
- the weighting factor calculation unit 16b calculates the weighting factor w2 by the equation (12) using the interframe distance tb and the interframe distance td calculated by the reference value calculation unit 16a.
- w2 clip3 (th1 ′, th2 ′, tb / td) (12)
- th1 ′ and th2 ′ are a lower limit threshold and an upper limit threshold of clip processing that change according to the partition size.
- the difference between the upper and lower thresholds for clip processing has a positive correlation with the partition size.
- weighting factor calculation unit 16b calculates the weighting factor w1 from the equation (9) using the calculated weighting factor w2.
- the optimum weighting factor tends to be different when the partition size is different.
- the optimum weighting coefficient is a weighting coefficient derived by the least square method so that the square error between the original image and the decoded image is minimized in the moving image decoding apparatus 1.
- the partition size when the partition size is large, the prediction error that may occur in the predicted image tends to be further reduced by bringing the ratio of the weight coefficient w1 and the weight coefficient w2 closer to the ratio of the interframe distance. On the other hand, when the partition size is small, the noise that can occur in the predicted image tends to be further reduced by bringing the ratio of the weight coefficient w1 and the weight coefficient w2 close to 1: 1.
- the ratio between the weighting factor w1 and the weighting factor w2 approaches the ratio of the interframe distances, so that a prediction error that may occur in the predicted image is increased.
- the ratio between the weighting factor w1 and the weighting factor w2 approaches 1: 1, so that noise can be reduced. Thereby, since a prediction image with few errors can be generated, coding efficiency can be improved.
- the lower limit threshold th1 ′ and the upper limit threshold th2 ′ of clip processing according to the partition size used in this example are not limited to these, and are optimal values for generating a predicted image with less errors. Can be set. If the partition size is large, a method may be used in which clipping is not performed.
- weighting factors w1 and w2 corresponding to the partition size are used as the weighting factors w1 and w2 corresponding to the partition size.
- r1 and r2 are rational coefficient coefficients common to the denominator that approximates the optimum weight coefficient.
- FIG. 5 shows optimum weighting factors determined by the least square method for each partition size.
- the coefficients r1 and r2 corresponding to each partition size can be set as follows, for example, based on the graph shown in FIG.
- (r1, r2) is set to (14/16, 2/16), and when the partition size is 16 ⁇ 8 or 8 ⁇ 16, (r1, r2 ) Is set to (10/16, 6/16) and the partition size is 8 ⁇ 8, (r1, r2) is set to (6/16, 10/16) and the partition size is 8 ⁇ 4 Alternatively, if 4 ⁇ 8, (r1, r2) is set to (4/16, 12/16), and if the partition size is 4 ⁇ 4, (r1, r2) is set to (2/16, 14 / 16).
- an error is modeled and an optimal weighting factor is obtained.
- mc1 and mc2 represent pixel values of the motion compensated image.
- the error E is differentiated with respect to w1.
- the weighting factors w1 and w2 such that the ratio of the weighting factors is 1: 1 are the optimum weighting factors. It can be said that there is.
- t1 and t2 represent a time difference from the prediction target image (decoded image being decoded).
- the error E is differentiated with respect to w1.
- the weighting factor w2 can be expressed by equation (16) by substituting equation (15) into equation (9).
- weighting factor w1 and w2 are represented by the ratio of the interframe distances, and it can be said that the weighting factor is the optimum weighting factor.
- the weighting coefficient used in AVC's implied_mode is represented by the ratio of the interframe distance between the reference frame and the predicted frame with respect to an image displayed with constant velocity motion.
- the optimum weighting factors w1 and w2 obtained using the equations (15) and (16) are H.264 / MPEG-4. It can be said that it is the same value as the weighting coefficients w1 and w2 obtained by AVC's implicit_mode.
- the optimum weighting factor is estimated as follows.
- the error E is differentiated with respect to w1.
- the weighting factor w2 can be expressed by equation (18) by substituting equation (17) into equation (9).
- the encoded data encoded using the above method can generate encoded data with high encoding efficiency, and generates a decoded image with less error with respect to the encoded data generated in this way. can do.
- the optimum weighting factor estimate when the error is composed of non-stationary time-varying components is as follows.
- E
- 2
- E
- 2 W1 2 ⁇ NC1 ⁇ t1 2 + (1-w1) 2 ⁇ NC2 ⁇ t2 2 Get.
- w1 (NC2 ⁇ t2 2 ) / (NC1 ⁇ t1 2 + NC2 ⁇ t2 2 ) (17 ′)
- the weighting factor w2 is expressed by the following equation (18 ′) using the equation (9).
- w2 (NC1 ⁇ t1 2 ) / (NC1 ⁇ t1 2 + NC2 ⁇ t2 2 ) (18 ′)
- nc1 and nc2 are substantially equal to each other, that is, when NC1 and NC2 are substantially equal to each other, w1 and w2 are expressed as in the following equations (17 ′′) and (18 ′′).
- the weighting factor calculation unit 16b may calculate the weighting factors w1 and w2 using the equations (17 ′), (18 ′), (17 ′′), and (18 ′′). Good.
- the video decoding device 1 may perform weighted prediction by a predetermined method, or may perform weighted prediction by a plurality of predetermined methods.
- the weighted prediction may be performed by a method selected from the methods, and the selected method may be explicitly encoded.
- each of the weighting methods described above can be used as one of selectable weighting methods.
- any of the weighting methods described above, H.264 / MPEG-4. A configuration in which the Explicit mode defined in the AVC is selected as a selection candidate, a mode in which the processing for making the weighting factor close to 0.5 as described above, and a mode in which such processing is not performed are selected as selection candidates. And so on.
- the moving image encoding apparatus 2 includes H.264 as a part thereof. H.264 / MPEG-4. It is a decoding device including a technique adopted in AVC and KTA software.
- FIG. 7 is a block diagram showing a configuration of the moving picture encoding apparatus 2.
- the moving image encoding device 2 includes a transform / quantization unit 21, a variable-length code encoding unit 22, an inverse quantization / inverse transform unit 23, a buffer memory 24, an intra-predicted image generation unit 25, An inter-predicted image generation unit 26, a motion vector detection unit 27, a prediction method control unit 28, a motion vector redundancy deletion unit 29, an adder 31, and a subtractor 32 are provided.
- the moving image encoding device 2 is a device that generates encoded data # 1 by encoding moving image # 10.
- the transform / quantization unit 21 performs (1) DCT transform (Discrete Cosine Transform) on the prediction residual D obtained by subtracting the predicted image Pred from the encoding target image, and (2) DCT coefficients obtained by the DCT transform. (3)
- the quantized prediction residual QD obtained by the quantization is supplied to the variable-length code encoding unit 22 and the inverse quantization / inverse transform unit 23.
- the transform / quantization unit 21 selects (1) a quantization step QP to be used for quantization for each macroblock, and (2) a quantization parameter difference ⁇ qp indicating the size of the selected quantization step QP. Is supplied to the variable-length code encoding unit 22, and (3) the selected quantization step QP is supplied to the inverse quantization / inverse transform unit 23.
- variable-length code encoding unit 22 performs variable-length encoding on the quantized prediction residuals QD and ⁇ qp supplied from the transform / quantization unit 21 together with the quantization parameter PP supplied from the prediction scheme control unit 28 described later. Thus, encoded data is generated.
- the inverse quantization / inverse transform unit 23 (1) inversely quantizes the quantized prediction residual QD, (2) performs inverse DCT (Discrete Cosine Transform) transformation on the DCT coefficient obtained by the inverse quantization, and (3) The prediction residual D obtained by the inverse DCT transform is supplied to the adder 31.
- the quantization step QP supplied from the transform / quantization unit 21 is used.
- the prediction residual D output from the inverse quantization / inverse transform unit 23 is obtained by adding a quantization error to the prediction residual D input to the transform / quantization unit 21. Common names are used for this purpose.
- the intra predicted image generation unit 25 generates a predicted image Pred_Intra related to each partition. Specifically, (1) a prediction mode used for intra prediction is selected, and (2) a predicted image Pred_Intra is generated from the local decoded image P using the selected prediction mode.
- the decoded image P is a local decoded image being decoded that has not yet been decoded
- the intra predicted image generation unit 25 applies the block that has already been decoded among the pixels constituting the local decoded image P.
- the pixel value of each pixel to which it belongs can be read from the buffer memory 24.
- the intra-predicted image generation unit 25 supplies the generated intra-predicted image Pred_Intra to the prediction method control unit 28 together with prediction mode designation information that designates the selected prediction mode.
- the motion vector detection unit 27 detects a motion vector mv related to each partition. Specifically, (1) selecting a decoded image P ′ to be used as a reference image, and (2) searching for a region that best approximates the target partition in the selected decoded image P ′, thereby moving the motion vector mv related to the control partition. Is detected.
- the decoded image P ′ is a decoded image that has already been decoded, and the motion vector detection unit 27 can read out the pixel value of each pixel constituting the decoded image P ′ from the buffer memory 24. .
- the motion vector detection unit 27 supplies the detected motion vector mv to the inter prediction image generation unit 26 and the motion vector redundancy deletion unit 29 together with the reference image index RI that specifies the decoded image P ′ used as the reference image.
- the reference image index RI specifies the decoded image P ′ used as the reference image.
- two decoded images P1 and P2 are selected as reference images, and motion vectors mv1 and mv2 corresponding to the two decoded images P1 and P2, respectively.
- the reference image indexes RI1 and RI2 are supplied to the inter prediction image generation unit 26 and the motion vector redundancy deletion unit 29.
- the inter prediction image generation unit 26 generates a motion compensation image mc related to each inter prediction partition. Specifically, the motion compensation image mc is generated from the decoded image P ′ specified by the reference image index RI supplied from the motion vector detection unit 27 using the motion vector mv supplied from the motion vector detection unit 27. . Similar to the motion vector detection unit 27, the inter predicted image generation unit 26 can read out the pixel value of each pixel constituting the decoded image P ′ from the buffer memory 24. The inter prediction image generation unit 26 supplies the generated motion compensated image mc (intra prediction image Pred_Intra) to the prediction method control unit 28 together with the reference image index RI supplied from the motion vector detection unit 27.
- intra prediction image Pred_Intra intra prediction image Pred_Intra
- the motion compensated image mc1 is generated from the decoded image P1 specified by the reference image index RI1 using the motion vector mv1, and (2) the motion vector Using mv2, a motion compensated image mc2 is generated from the reference image P2 specified by the reference image index RI2, and (3) an offset value is added to the weighted average of the motion compensated image mc1 and the motion compensated image mc2.
- a predicted image Pred_Inter is generated.
- the prediction method control unit 28 compares the intra predicted image Pred_Intra and the inter predicted image Pred_Inter with the encoding target image, and selects whether to perform intra prediction or inter prediction.
- the prediction scheme control unit 28 supplies the intra prediction image Pred_Intra as the prediction image Pred to the adder 31 and the subtracter 32, and also uses the prediction mode designation information supplied from the intra prediction image generation unit 25.
- the prediction parameter PP is supplied to the variable length code encoder 22.
- the prediction scheme control unit 28 supplies the inter prediction image Pred_Inter as the prediction image Pred to the adder 31 and the subtractor 32, and the reference image index supplied from the inter prediction image generation unit 26.
- the RI and the estimated motion vector index PMVI and the motion vector residual MVD supplied from the motion vector redundancy deleting unit 29 (described later) are supplied to the variable length code encoding unit as the prediction parameters PP.
- the prediction residual D is generated by the subtracter 32 by subtracting the prediction image Pred selected by the prediction method control unit 28 from the encoding target image.
- the prediction residual D generated by the subtractor 32 is DCT transformed / quantized by the transform / quantization unit 21 as described above.
- the adder 31 by adding the prediction image Pred selected by the prediction method control unit 28 to the prediction residual D generated by the inverse quantization / inverse conversion unit 23, the adder 31 generates a local decoded image P. Generated.
- the local decoded image P generated by the adder 31 is stored in the buffer memory 24 and used as a reference image in inter prediction.
- the motion vector redundancy deleting unit 29 deletes the redundancy in the motion vector mv detected by the motion vector detecting unit 27. Specifically, (1) an estimation method used for estimating the motion vector mv is selected, (2) an estimated motion vector pmv is derived according to the selected estimation method, and (3) the estimated motion vector pmv is subtracted from the motion vector mv. As a result, a motion vector residual MVD is generated.
- the motion vector redundancy deletion unit 29 supplies the generated motion vector residual MVD to the prediction method control unit 28 together with the estimated motion vector index PMVI indicating the selected estimation method.
- FIG. 8 is a block diagram illustrating a configuration of the inter predicted image generation unit 26.
- the inter prediction image generation unit 26 includes a reference value calculation unit 26a, a weight coefficient calculation unit 26b, a motion compensation unit 26c, and a prediction image generation unit 26d.
- the reference value calculation unit 26a, the weight coefficient calculation unit 26b, the motion compensation unit 26c, and the prediction image generation unit 26d included in the inter prediction image generation unit 26 of the moving image encoding device 2 are respectively connected to the moving image decoding device 1.
- the inter-predicted image generation unit 16 has the same functions as the reference value calculation unit 16a, the weight coefficient calculation unit 16b, the motion compensation unit 16c, and the predicted image generation unit 16d.
- the reference picture indexes RI and R2 and the motion vectors mv1 and mv2 are supplied from the motion vector restoring unit 14, whereas in the moving picture encoding apparatus 2, these are the motion vector detection. Supplied from the unit 27.
- the POC designation information and the reference images L 0 and L 1 are read from the buffer memory 15, whereas in the video encoding device 2, these are read from the buffer memory 24.
- weighting coefficient calculation unit 26b performs any one of the calculation processes listed in the following (specific example 1-1 ′) to (specific example 1-5 ′).
- FIG. 9 shows the code amount reduction rate for each sequence in the encoded data # 1 when encoding is performed using the inter-predicted image Pred_Inter generated using the weight coefficients w1 and w2 calculated in this way. .
- FIG. 9 shows encoded data # 1 generated using the weighting factor of this example, and H.264. H.264 / MPEG-4. It is a graph which shows the code amount reduction rate for every sequence in the coding data produced
- the moving picture encoding apparatus 2 can reduce the code amount of the encoded data # 1 by using this example.
- the moving image coding apparatus 2 Since the inter prediction image Pred_Inter generated using the weighting factors w1 and w2 calculated in this way has reduced errors due to noise, the moving image coding apparatus 2 performs coding with a smaller code amount. Data # 1 can be generated.
- the inter prediction image Pred_Inter generated using the weighting factors w1 and w2 calculated in this way has a smaller prediction error when the partition size is large, and further reduces an error due to noise when the partition size is small.
- the moving picture encoding apparatus 2 can generate encoded data # 1 with a smaller code amount.
- the inter prediction image Pred_Inter generated using the weighting factors w1 and w2 calculated in this way has a smaller prediction error when the partition size is large, and further reduces an error due to noise when the partition size is small.
- the moving picture encoding apparatus 2 can generate encoded data # 1 with a smaller code amount.
- the encoded data encoded using the above method can generate encoded data with high encoding efficiency.
- FIG. 10 is a block diagram illustrating a configuration of the moving image decoding apparatus 3.
- the same blocks as those shown in FIG. 1 are denoted by the same reference numerals, and the description thereof is omitted.
- the video decoding device 3 includes a variable length code decoding unit 13, a motion vector restoration unit 14, a buffer memory 15, an inter prediction image generation unit 16 ′, an intra prediction image generation unit 17, and a prediction method determination unit. 18, an inverse quantization / inverse transform unit 19 and an adder 20 are provided.
- the difference between the moving image decoding device 3 (the present embodiment) and the moving image decoding device 1 (the first embodiment) is that weights w1 and w2 for weighted prediction based on the POC designation information stored in the buffer memory 15.
- the inter-predicted image generation unit 16 ′ that sets the weights w1 ′ and w2 ′ of the weighted prediction based on the motion vector mv restored by the motion vector restoration unit 14 is provided instead of the inter-predicted image generation unit 16 that sets It is a point.
- the inter predicted image generation unit 16 ′ will be described in more detail as follows.
- FIG. 11 is a block diagram illustrating a configuration of the inter predicted image generation unit 16 ′.
- the inter predicted image generation unit 16 ′ includes a reference value calculation unit 16a ′, a weight coefficient calculation unit 16b ′, a motion compensation unit 16c, and a predicted image generation unit 16d.
- the inter-predicted image generation unit 16 ′ can perform unidirectional prediction and bi-directional prediction.
- a configuration for performing bi-directional prediction weighted prediction
- the reference value calculation unit 16a reads out the motion vectors mv1 and mv2 associated with the reference image indexes RI1 and RI2 from the buffer memory 15, respectively. Then, the size ml1 of the motion vector mv1 and the size ml2 of the motion vector mv2 are calculated (hereinafter, the size ml1 of the motion vector mv1 is also referred to as “motion vector length ml1”, and the size ml2 of the motion vector mv2) Is also referred to as “motion vector length ml2”). The calculated motion vector length ml1 and motion vector length ml2 are supplied to the weight coefficient calculation unit 16b '. A specific method for calculating the motion vector length ml1 and the motion vector length ml2 will be described later.
- the weighting factor calculation unit 16b ' calculates the weighting factors w1' and w2 'from the motion vector lengths ml1 and ml2 calculated by the reference value calculation unit 16a', and supplies them to the predicted image generation unit 16d.
- a method for calculating the weighting factors w1 'and w2' will be described later.
- the motion compensation unit 16c reads the reference images L0 (decoded image P1) and L1 (decoded image P2) designated by the reference image indexes RI1 and RI2 from the buffer memory 15. Then, motion compensation images mc1 and mc2 are generated from each of the reference images L1 and L0 by motion compensation using each of the motion vectors mv1 and mv2, and these are supplied to the predicted image generation unit 16d. Since the method of generating the motion compensated images mc1 and mc2 by the motion compensation unit 16c has already been described, description thereof is omitted here.
- the predicted image generation unit 16d uses the weighting coefficients w1 ′ and w2 ′ calculated by the weighting coefficient calculation unit 16b ′ and the motion compensation images mc1 and mc2 generated by the motion compensation unit 16c to generate an inter prediction image. Generate Pred_Inter and output it.
- FIG. 12 is a diagram illustrating an example of a motion vector mv1 used for motion compensation with reference to the reference image L0 and a motion vector mv2 used for motion compensation with reference to the reference image L1.
- the reference value calculation unit 16a calculates a motion vector length ml1 that is the magnitude of the motion vector mv1 by the following equation (19).
- abst Represents taking the absolute value of the vector in parentheses
- mv1x and mv1y represent the x component and the y component of the motion vector mv1, respectively.
- the reference value calculation unit 16a calculates the motion vector length ml2, which is the magnitude of the motion vector mv2, by the following equation (20).
- mv2x and mv2y represent the x component and the y component of the motion vector mv2, respectively.
- the motion vector length may be calculated by a method other than the above equations (19) and (20), for example, the sum of the absolute value of the x component and the absolute value of the y component.
- the weight coefficient calculation unit 16b ' calculates the weight coefficients w1' and w2 'used in the weighted prediction using the motion vector lengths ml1 and ml2 calculated by the reference value calculation unit 16a'. Specific examples of the calculation processing of the weighting factors w1 'and w2' will be described in (Specific Example 2-1) to (Specific Example 2-3) shown below.
- the predicted image generating unit 16d A pixel value pred (x, y) of each pixel of Pred_Inter is generated using Expression (7 ′).
- the weighting coefficient calculation unit 16b ′ performs any one of the calculation processes listed in the following (Specific example 2-1) to (Specific example 2-3), and each motion compensated image in Expression (7 ′). Weight coefficients w1 ′ and w2 ′ to be multiplied by each pixel value are calculated.
- the magnitudes ml1 and ml2 of the motion vectors mv1 and mv2 used for motion compensation for obtaining the motion compensated images mc1 and mc2 are ml1 ⁇ ml2
- the values of the weight coefficients w1 ′ and w2 ′ Becomes
- the values of the weighting factors w1 ′ and w2 ′ are set to predetermined values. Set otherwise, set to motion vector length ratio.
- th1 and th2 represent a lower limit threshold and an upper limit threshold in clip processing, and are set to values that can calculate optimum weighting factors w1 ′ and w2 ′ with high coding efficiency.
- the weight vector calculation unit 16b ′ performs the operation of the present example and uses the motion vector length ratio as the weight coefficient, the prediction accuracy is improved and the motion vector length ratio greatly deviates from 0.5.
- the increase in noise can be suppressed by performing the clipping process described above.
- the moving picture encoding apparatus provided with the same means as the weighting coefficient calculation unit 16b 'that performs the operation of this example can generate encoded data with high encoding efficiency.
- the moving picture decoding apparatus 3 including the weighting coefficient calculation unit 16b 'that performs the operation of the present example can appropriately decode the encoded data generated in such a manner with high encoding efficiency.
- the value of the weighting factor w2 ' is set by comparing the motion vector length ml1 and the motion vector length ml2.
- the weighting factor calculation unit 16b ' sets the weighting factor w2' using the motion vector lengths ml1 and ml2 calculated by the reference value calculation unit 16a '.
- the weighting factor calculation unit 16b calculates the weighting factor w1' from the equation (22) using the weighting factor w2 'set to one of the values (1) to (3).
- the weight is calculated based on the motion vector length only when the ratio of the distance between frames is 1: 1.
- a reference value calculation unit 16a included in the video decoding device 1 according to the first embodiment is added to the inter predicted image generation unit 16 ′, and the calculation is performed by the reference value calculation unit 16a. It is preferable to adopt a configuration in which the interframe distances tb and td are supplied to the weighting coefficient calculation unit 16b ′.
- the weighting factor calculation unit 16b ' calculates the weighting factor w2 based on the motion vector length ml1, the motion vector length ml2, the interframe distance tb, and the interframe distance td calculated by the reference value calculation unit 16a'.
- the weight coefficient w2 ′ is calculated by the following equation (23).
- w2 ′ ml1 / (ml1 + ml2) (23)
- weighting factor calculation unit 16b calculates the weighting factor w1' by using the calculated weighting factor w2 'from the equation (22).
- tb and the inter-frame distance td are also output to the weighting factor calculation unit 16b ′, this example is not limited to this.
- FIG. 13 is a block diagram illustrating a configuration of the moving image encoding device 4.
- the same reference numerals are assigned to the same blocks as those shown in FIG. 7, and the description thereof is omitted.
- the moving image encoding device 4 includes a transform / quantization unit 21, a variable-length code encoding unit 22, an inverse quantization / inverse transform unit 23, a buffer memory 24, an intra predicted image generation unit 25, An inter-predicted image generation unit 26 ′, a motion vector detection unit 27, a prediction method control unit 28, a motion vector redundancy deletion unit 29, an adder 31, and a subtracter 32 are provided.
- the moving image encoding device 4 is a device that generates encoded data # 3 by encoding moving image # 30.
- the difference between the moving image encoding device 4 (the present embodiment) and the moving image encoding device 2 (the first embodiment) is that the weight w1 of the weighted prediction based on the POC designation information stored in the buffer memory 24.
- the inter prediction image generation unit 26 ′ that sets the weights w1 ′ and w2 ′ of the weighted prediction based on the motion vector mv detected by the motion vector detection unit 27. It is a point equipped with.
- FIG. 14 is a block diagram illustrating a configuration of the inter predicted image generation unit 26 ′.
- the inter predicted image generation unit 26 ′ includes a reference value calculation unit 26a ′, a weight coefficient calculation unit 26b ′, a motion compensation unit 26c, and a predicted image generation unit 26d.
- the reference value calculation unit 26a ′, the weight coefficient calculation unit 26b ′, the motion compensation unit 26c, and the prediction image generation unit 26d included in the inter prediction image generation unit 26 ′ of the moving image encoding device 4 are each a moving image.
- the inter prediction image generation unit 16 ′ of the decoding device 3 has functions equivalent to those of the reference value calculation unit 16a ′, the weight coefficient calculation unit 16b ′, the motion compensation unit 16c, and the prediction image generation unit 16d. .
- the reference image indexes RI and R2 and the motion vectors mv1 and mv2 are supplied from the motion vector restoration unit 14, whereas in the moving image encoding device 4, these are detected as motion vector detection. Supplied from the unit 27.
- the reference images L 0 and L 1 are read from the buffer memory 15, whereas in the video encoding device 4, these are read from the buffer memory 24.
- weighting factor calculation unit 26b ′ performs any one of the calculation processes listed in the following (specific example 2-1 ′) to (specific example 2-3 ′).
- the moving image encoding device 4 Since the inter prediction image Pred_Inter generated using the weighting factors w1 ′ and w2 ′ calculated in this way has reduced errors due to noise, the moving image encoding device 4 has a smaller code amount. Encoded data # 3 can be generated.
- the moving picture encoding apparatus 4 can reduce the code amount of the encoded data # 3 by using this example.
- the weighting factor w2 ′ is the motion vector length ml1 in the same manner as in the calculation processing of the weighting factors w1 ′ and w2 ′ by the weighting factor calculating unit 16b ′ of the video decoding device 3 (specific example 2-2).
- ml2 are set to any one of (1) to (3) described in (Specific Example 2-2).
- weighting factor w1 ' is calculated from the equation (22) using the weighting factor w2' as in (Specific Example 2-2).
- the moving image encoding device 4 Since the inter prediction image Pred_Inter generated using the weighting factors w1 ′ and w2 ′ calculated in this way has reduced errors due to noise, the moving image encoding device 4 has a smaller code amount. Encoded data # 3 can be generated.
- weighting factor w1 ' is calculated from the equation (22) using the weighting factor w2' as in (Specific Example 2-3).
- the moving image encoding device 4 can generate encoded data # 3 with a smaller code amount.
- FIG. 15 is a block diagram illustrating a configuration of the moving image decoding apparatus 5.
- the same blocks as those shown in FIG. 1 are denoted by the same reference numerals, and the description thereof is omitted.
- the moving image decoding apparatus 5 includes a variable length code decoding unit 13, a motion vector restoration unit 14, a buffer memory 15, an inter prediction image generation unit 16 ′′, an intra prediction image generation unit 17, and a prediction method determination unit. 18, an inverse quantization / inverse transform unit 19 and an adder 20 are provided.
- the first difference between the moving picture decoding apparatus 5 (the present embodiment) and the moving picture decoding apparatus 1 (the first embodiment) is that the inverse quantization / inverse transform unit 19 is decoding the quantization parameter qp. The point is that it is stored in the buffer memory 15 in association with the image P.
- the second difference between the moving picture decoding apparatus 5 (the present embodiment) and the moving picture decoding apparatus 1 (the first embodiment) is that weighting is performed by referring to the POC designation information stored in the buffer memory 15.
- the weighting prediction weight w1 by referring to the quantization parameter qp ′ stored in the buffer memory 15 by the inverse quantization / inverse transformation unit 19
- the inter-predicted image generating unit 16 ′′ for setting “and w2” is provided.
- the inter-predicted image generating unit 16 ′′ will be described in more detail as follows.
- FIG. 16 is a block diagram illustrating the configuration of the inter predicted image generation unit 16 ′′.
- the inter predicted image generation unit 16 ′′ includes a reference value calculation unit 16a ′′, a weight coefficient calculation unit 16b ′′, a motion compensation unit 16c, and a prediction image generation unit 16d.
- the predicted image generation unit 16 ′′ can perform unidirectional prediction and bidirectional prediction. A configuration for performing bidirectional prediction (weighted prediction) will be described below.
- the reference value calculation unit 16a ′′ reads the quantization parameters qp1 and qp2 associated with the decoded images P1 and P2 (hereinafter referred to as reference images L0 and L1) designated by the reference image indexes RI1 and RI2 from the buffer memory 15. Then, each of the quantization steps QP1 and QP2 used when decoding the reference images L0 and L1 is calculated from the read quantization parameters qp1 and qp2, and a specific calculation method for the quantization steps QP1 and QP2 is calculated. Will be described later.
- the weighting factor calculation unit 16b ′′ calculates the weighting factors w1 ′′ and w2 ′′ from the quantization steps QP1 and QP2 calculated by the reference value calculation unit 16a ′′, and supplies these to the predicted image generation unit 16d. A method for calculating the weighting factors w1 "and w2" will be described later.
- the motion compensation unit 16c reads the reference images L0 (decoded image P1) and L1 (decoded image P2) designated by the reference image indexes RI1 and RI2 from the buffer memory 15. Then, motion compensation images mc1 and mc2 are generated from each of the reference images L1 and L0 by motion compensation using each of the motion vectors mv1 and mv2, and these are supplied to the predicted image generation unit 16d. Since the method of generating the motion compensated images mc1 and mc2 by the motion compensation unit 16c has already been described, description thereof is omitted here.
- the prediction image generation unit 16d generates the inter prediction image Pred_Inter using the weight coefficients w1 and w2 calculated by the weight coefficient calculation unit 16b and the motion compensation images mc1 and mc2 generated by the motion compensation unit 16c. And output this.
- FIG. 17 illustrates the motion vector mv1 and the reference image used for motion compensation with reference to the reference image L0. It is a figure showing an example of the motion vector mv2 used for the motion compensation with reference to L1.
- the reference value calculation unit 16a ′′ calculates the quantization step QP1 by the following equation (24).
- qp1 / 6 represents a quotient obtained by dividing qp1 by 6.
- the reference value calculation unit 16a ′′ calculates the quantization step QP1 by the following equation (25).
- qp2 / 6 represents a quotient obtained by dividing qp2 by 6.
- the weighting factor calculation unit 16b ′′ calculates the weighting factors w1 ′′ and w2 ′′ using the quantization steps QP1 and QP2 calculated by the reference value calculation unit 16a ′′. Specific examples of the calculation processing of the weighting factors w1 ′′ and w2 ′′ will be described in the following (Specific Example 3-1) to (Specific Example 3-3).
- the predicted image generation unit 16d generates the pixel value pred (x, y) of each pixel of the inter predicted image Pred_Inter using the equation (7 ′′).
- the weighting coefficient calculation unit 16b ′′ performs any one of the calculation processes listed in the following (Specific example 3-1) to (Specific example 3-3), and performs the calculation of each motion compensation image in Expression (7).
- Weight coefficients w1 "and w2" multiplied by each pixel value are calculated. In any of the specific examples, they are used for inverse quantization to obtain each of the reference images L0 (decoded image P1) and L1 (decoded image P2).
- the quantization steps QP 1 and QP 2 satisfy QP 1 ⁇ QP 2 , the values of the weighting factors w 1 ′′ and w 2 ′′ are
- the weighting factor calculation unit 16b ′′ uses the quantization steps QP1 and QP2 calculated by the reference value calculation unit 16a ′′ to calculate the weighting factors w1 ′′ and w2 ′′ according to equations (26) and (27).
- th1 and th2 represent a lower limit threshold and an upper limit threshold in clip processing, and are set to values that can calculate optimum weighting factors w1 ′′ and w2 ′′ with high encoding efficiency.
- the weighting factor calculation unit 16b ′′ performs the operation of this example and uses the quantization step ratio as the weighting factor, the prediction step accuracy is improved and the quantization step ratio greatly deviates from 0.5.
- the increase in noise can be suppressed by performing the clipping process described above.
- the weighting factor calculation unit 16b ′′ calculates the weighting factor w2 ′′ using the quantization steps QP1 and QP2.
- the quantization step QP1 is smaller than the value obtained by subtracting 6 from the quantization step QP2 (
- ⁇ 6), w2 ′′ th1 ′ and (2) quantum The quantization step QP1 is equal to or greater than the value obtained by subtracting 6 from the quantization step QP2, and the quantization step QP1 is equal to or less than the value obtained by adding 6 to the quantization step QP2 ((
- >
- ⁇
- +6)), w2 " 0.5, and (3) the quantization step QP1 is larger than the value obtained by adding 6 to the quantization step QP2 (
- +6), w2 ′′ th2 ′.
- the weighting factor calculation unit 16b ′′ calculates the weighting factor w1 from the formula (27) using the weighting factor w2 ′′ set to any one of the above (1) to (3).
- the weight is calculated based on the quantization step.
- a reference value calculation unit 16a included in the video decoding device 1 according to the first embodiment is added to the inter predicted image generation unit 16 ′, and the calculation is performed by the reference value calculation unit 16a. It is preferable to adopt a configuration in which the interframe distances tb and td are supplied to the weighting coefficient calculation unit 16b ′.
- the weighting factor calculation unit 16b ′′ calculates the weighting factor w2 ′′ using the quantization step QP1, the quantization step QP2, the interframe distance tb, and the interframe distance td calculated by the reference value calculation unit 16a ′′. .
- the value calculated in the first embodiment is also preferable. It is.
- the weighting factor calculation unit 16b ′′ calculates the weighting factor w1 from the formula (27) using the calculated weighting factor w2 ′′.
- FIG. 18 is a block diagram illustrating a configuration of the moving image encoding device 6.
- the same reference numerals are assigned to the same partitions as those shown in FIG. 7, and the description thereof is omitted.
- the moving image encoding device 6 includes a transform / quantization unit 21, a variable length code encoding unit 22, an inverse quantization / inverse transform unit 23, a buffer memory 24, an intra predicted image generation unit 25, An inter-predicted image generation unit 26 ′′, a motion vector detection unit 27, a prediction method control unit 28, a motion vector redundancy deletion unit 29, an adder 31, and a subtractor 32 are provided.
- the first difference between the moving image encoding device 6 (the present embodiment) and the moving image encoding device 2 (the first embodiment) is that the transform / quantization unit 21 is decoding the quantization parameter qp.
- the point is that the image is stored in the buffer memory 24 in association with the image P.
- the second difference between the moving image encoding device 6 (the present embodiment) and the moving image encoding device 2 (the first embodiment) is that the POC designation information stored in the buffer memory 24 is referred to.
- the weighting prediction weight w1 by referring to the quantization parameter qp ′ stored in the buffer memory 24 by the transform / quantization unit 21
- the inter prediction image generation unit 26 ′′ for setting “and w2” is provided.
- the inter prediction image generation unit 26 ′′ will be specifically described as follows.
- FIG. 19 is a block diagram illustrating a configuration of the inter predicted image generation unit 26 ′′.
- the inter predicted image generation unit 26 ′′ includes a reference value calculation unit 26a ′′, a weight coefficient calculation unit 26b ′′, A motion compensation unit 26c and a predicted image generation unit 26d are provided.
- the reference value calculation unit 26a ′′, the weight coefficient calculation unit 26b ′′, the motion compensation unit 26c, and the prediction image generation unit 26d included in the inter prediction image generation unit 26 ′′ of the moving image encoding device 6 are each a moving image.
- the inter prediction image generation unit 16 ′′ of the decoding device 5 has the same functions as the reference value calculation unit 16a ′′, the weight coefficient calculation unit 16b ′′, the motion compensation unit 16c, and the prediction image generation unit 16d. .
- the reference image indexes RI and R2 and the motion vectors mv1 and mv2 are supplied from the motion vector restoring unit 14, whereas in the moving image encoding device 6, these are detected as motion vector detection. Supplied from the unit 27.
- the quantization parameters qp1 and qp2 and the reference pictures L0 and L1 are read from the buffer memory 15, whereas in the moving picture encoding apparatus 4, these are read from the buffer memory 24. .
- the weighting factor calculating unit 26b ′′ includes the following (specific examples 3-1 ′) to (specific examples). One of the calculation processes listed in 3-3 ′) is performed.
- the moving image coding apparatus 6 Since the inter-predicted image Pred_Inter generated using the weighting factors w1 ′′ and w2 ′′ calculated in this way has reduced errors due to coding noise, the moving image coding apparatus 6 has a smaller code amount. It is possible to generate encoded data # 5 with a small amount of data.
- the moving picture encoding apparatus 6 can reduce the code amount of the encoded data # 5 by using this example.
- weighting factor w1 ′′ is calculated from the equation (27) using the weighting factor w2 ′′ as in (Specific Example 3-2).
- the moving image coding apparatus 6 Since the inter-predicted image Pred_Inter generated using the weighting factors w1 ′′ and w2 ′′ calculated in this way has reduced errors due to coding noise, the moving image coding apparatus 6 has a smaller code amount. It is possible to generate encoded data # 5 with a small amount of data.
- the weighting factor w1 ′′ is calculated from the equation (27) using the weighting factor w2 ′′ as in (Specific Example 3-3).
- the moving image encoding device 6 can generate encoded data # 5 with a smaller code amount.
- each block of the above-described moving picture decoding apparatuses 1, 3, 5 and moving picture coding apparatuses 2, 4, 6 is realized in hardware by a logic circuit formed on an integrated circuit (IC chip). Alternatively, it may be realized by software using a CPU (Central Processing Unit).
- IC chip integrated circuit
- CPU Central Processing Unit
- each device includes a CPU that executes instructions of a program that realizes each function, a ROM (Read (Memory) that stores the program, a RAM (Random Memory) that expands the program, the program, and various types
- a storage device such as a memory for storing data is provided.
- An object of the present invention is to provide a recording medium in which a program code (execution format program, intermediate code program, source program) of a control program for each of the above devices, which is software that realizes the above-described functions, is recorded in a computer-readable manner. This can also be achieved by supplying each of the above devices and reading and executing the program code recorded on the recording medium by the computer (or CPU or MPU).
- Examples of the recording medium include tapes such as magnetic tapes and cassette tapes, magnetic disks such as floppy (registered trademark) disks / hard disks, and disks including optical disks such as CD-ROM / MO / MD / DVD / CD-R.
- IC cards including memory cards
- semiconductor memories such as mask ROM / EPROM / EEPROM / flash ROM, or PLD (Programmable logic device) or FPGA (Field Programmable Gate Array) Logic circuits can be used.
- each of the above devices may be configured to be connectable to a communication network, and the program code may be supplied via the communication network.
- the communication network is not particularly limited as long as it can transmit the program code.
- the Internet intranet, extranet, LAN, ISDN, VAN, CATV communication network, virtual private network (Virtual Private Network), telephone line network, mobile communication network, satellite communication network, etc. can be used.
- the transmission medium constituting the communication network may be any medium that can transmit the program code, and is not limited to a specific configuration or type.
- wired lines such as IEEE 1394, USB, power line carrier, cable TV line, telephone line, ADSL (Asymmetric Digital Subscriber Line) line, infrared rays such as IrDA and remote control, Bluetooth (registered trademark), IEEE 802.11 wireless, HDR ( It can also be used by wireless such as High Data Rate, NFC (Near Field Communication), DLNA (Digital Living Network Alliance), mobile phone network, satellite line, terrestrial digital network.
- wired lines such as IEEE 1394, USB, power line carrier, cable TV line, telephone line, ADSL (Asymmetric Digital Subscriber Line) line, infrared rays such as IrDA and remote control, Bluetooth (registered trademark), IEEE 802.11 wireless, HDR ( It can also be used by wireless such as High Data Rate, NFC (Near Field Communication), DLNA (Digital Living Network Alliance), mobile phone network, satellite line, terrestrial digital network.
- the predicted image generation apparatus uses motion-compensated image generation means for generating motion-compensated images mc 1 and mc 2 for the decoded image P using each of the decoded images P 1 and P 2 as a reference image.
- Weight coefficient setting means for setting values of weight coefficients w 1 and w 2 to be multiplied by the motion compensated images mc 1 and mc 2 generated by the motion compensated image generation means, and the weight coefficient setting means By adding an offset value to the weighted average w 1 ⁇ mc 1 + w 2 ⁇ mc 2 of the motion compensated images mc 1 and mc 2 generated by the motion compensated image generating means using the weighting factors w 1 and w 2 And a predicted image generation unit that generates a predicted image Pred for the decoded image P.
- the predicted image generation method includes a motion compensated image generation step for generating motion compensated images mc 1 and mc 2 for the decoded image P using each of the decoded images P 1 and P 2 as a reference image, and the motion described above.
- a weighting factor setting step of setting the value of the weight coefficients w 1 and w 2 to be multiplied to each of the motion generated by compensated image generation step compensated image mc 1 and mc 2, set by the weighting factor setting step weights By adding the offset value to the weighted average w 1 ⁇ mc 1 + w 2 ⁇ mc 2 of the motion compensation images mc 1 and mc 2 generated in the motion compensation image generation step using the coefficients w 1 and w 2 .
- a predicted image generation step of generating a predicted image Pred for the decoded image P.
- the predicted image generation apparatus uses the weight coefficient setting means to set the inter-frame distance between the decoded image P 1 and the decoded image P 2 to be td, and the decoded image P 1 and the decoded image.
- the distance from P is tb, and the value of the weighting factor w 2 is set to satisfy 0 ⁇
- the predicted image generation method sets the interframe distance between the decoded image P 1 and the decoded image P 2 to be td and the decoded image P 1 and the decoded image in the weight coefficient setting step.
- the distance from the image P is tb, and the weight coefficient w 2 is set so that 0 ⁇
- the weighting factor w 2 is set so that 0 ⁇
- the weighting coefficient w 2 finally multiplied by the motion compensated image mc 2 satisfies the condition 0 ⁇
- the pixel value of the predicted image is obtained by dividing the pixel value obtained by adding the offset value to the weighted average W 1 ⁇ mc 1 + W 2 ⁇ mc 2 using the integer coefficients W 1 and W 2 by the constant C.
- the weight coefficient w 2 W 2 / C only needs to satisfy the condition 0 ⁇
- the offset value may be set to 0. That is, a configuration in which the offset value is not substantially added is also included in the scope of the present invention.
- the weighting factor setting means is, for example, used to set the value of the weight factor w 2, the clip3 (Th1, Th2, tb / td) using the threshold Th1 and Th2 (Th1 ⁇ 0.5 ⁇ Th2)
- the weight coefficient w 2 may be set to (tb / td ⁇ 0.5) ⁇ + 0.5 using the coefficient ⁇ (0 ⁇ ⁇ 1), or the coefficient ⁇ (0 ⁇ It may be set to clip3 (Th1, Th2, (tb / td ⁇ 0.5) ⁇ + 0.5) using ⁇ ⁇ 1) and threshold values Th1 and Th2 (Th1 ⁇ 0.5 ⁇ Th2). In either configuration, by bringing the value of the weight coefficient w 2 to 0.5, it is possible to reduce coding noise included in the predicted image.
- the weight coefficient setting means preferably sets the coefficient ⁇ in accordance with the size of the partition for which the predicted image Pred is to be generated.
- the weighting factor setting means preferably sets the threshold values Th1 and Th2 according to the size of the partition for which the predicted image Pred is to be generated.
- the optimum weighting factor tends to be different depending on the size of the partition for which the predicted image Pred is to be generated.
- the optimum weight coefficient is a weight coefficient derived by the least square method so that the square error between the encoding target image and the decoded image is minimized. For this reason, with the above configuration, the values of the weighting factors w 1 and w 2 can be made closer to the optimum weighting factor.
- the motion compensation image mc 1 and the motion vector mv 1 and mv 2 used for motion compensation to obtain mc 2
- the values of the weighting factors w 1 and w 2 are set so that
- the predicted image generation method uses the motion vectors mv 1 and mv used in motion compensation for obtaining the motion compensated images mc 1 and mc 2 in the weighting factor setting step.
- of 2 are
- the values of the weighting factors w 1 and w 2 are set to
- of the motion vectors mv 1 and mv 2 are
- the difference between the motion compensated image mc2 and the encoding target image It means that it is larger than the difference between the image mc1 and the encoding target image. Therefore, when the values of the weighting factors w 1 and w 2 are set to be
- Prediction accuracy is higher than setting to be. For this reason, when encoding the prediction residual which subtracted the prediction image from the encoding target image, there is an effect of reducing the possibility that the code amount increases due to the prediction.
- the weight coefficient setting means sets the value of the weight coefficient w 2 to
- ⁇ Th2) may be used to set to clip3 (Th1, Th2,
- 2 is set to a value smaller than 0.5 when (1)
- weighting factor setting means only when the inter-frame distance between the decoded image P 1 and the decoded image P is equal to the inter-frame distance between the decoded image P 2 and the decoded image P, weighting factors w 1 and w 2 May be set according to the magnitudes
- the predicted image generation apparatus uses the weighting factor setting unit to perform quantization steps QP 1 used for inverse quantization for obtaining each of the decoded images P 1 and P 2 , and
- QP 2 is QP 1 ⁇ QP 2
- the weight coefficients w 1 and w 2 are set so that
- the predicted image generation method is a quantization step QP used in inverse quantization for obtaining each of the decoded images P 1 and P 2 in the weighting factor setting step.
- QP 2 is QP 1 ⁇ QP 2
- set to become it is characterized in that.
- the fact that the quantization steps QP 1 and QP 2 satisfy QP 1 ⁇ QP 2 means that the coding noise (quantization noise) included in the motion compensated image mc2 is the coding noise (quantization noise) included in the motion compensated image mc1. ) Means greater than. Therefore, when the values of the weighting factors w 1 and w 2 are set to be
- the encoding noise included in the predicted image is smaller than the setting so as to be. For this reason, when encoding the prediction residual which subtracted the prediction image from the encoding target image, there is an effect of reducing the possibility that the code amount increases due to the prediction.
- the weighting factor setting means the value of the weight factor w 2, either set to QP 1 / (QP 1 + QP 2), or, the threshold Th1 and Th2 (Th1 ⁇ 0.5 ⁇ Th2) with clip3 (Th1, Th2, QP 1 / (QP 1 + QP 2 )) may be set, and k is a positive constant, and the value of the weighting factor w 2 is (1) QP 1 ⁇ QP 2 ⁇ Set to a value smaller than 0.5 when k, (2) Set to 0.5 when QP 2 ⁇ k ⁇ QP 1 ⁇ QP 2 + k, and (3) 0 when QP 2 + k ⁇ QP 1 A value larger than 5 may be set.
- the quantization step QP values of the weighting coefficients w 1 and w 2 1 and QP 2 may be set. In any configuration, the quantization noise included in the predicted image can be reduced.
- a moving picture decoding apparatus and a moving picture encoding apparatus provided with the predicted image generation apparatus are also included in the scope of the present invention.
- the present invention can be suitably used for a predicted image generation apparatus for generating a predicted image used in moving image predictive coding. Moreover, it can use suitably for a moving image decoding apparatus and a moving image encoding apparatus provided with such a predicted image generation apparatus.
Landscapes
- Engineering & Computer Science (AREA)
- Multimedia (AREA)
- Signal Processing (AREA)
- Compression Or Coding Systems Of Tv Signals (AREA)
Abstract
Description
(符号化データの構成)
本実施形態に係る動画像符号化装置2及び動画像復号装置1の説明に先立って、動画像符号化装置2によって生成され、動画像復号装置1によって復号される符号化データのデータ構造について説明を行う。動画像符号化装置2によって生成され、動画像復号装置1によって復号される符号化データは、シーケンスレイヤ、GOP(Group Of Pictures)レイヤ、ピクチャレイヤ、スライスレイヤ、及び、マクロブロックレイヤ、からなる階層構造を有している。
以下では、本実施形態に係る動画像復号装置1について図1~図6を参照して説明する。動画像復号装置1は、その一部に、H.264/MPEG-4.AVC、及び、KTAソフトウェアに採用されている技術を含む復号装置である。
次に、インター予測画像生成部16の構成について、図3を参照して説明する。図3は、インター予測画像生成部16の構成を示すブロック図である。図3に示すように、インター予測画像生成部16は、参照値算出部16a、重み係数算出部16b、動き補償部16c、及び、予測画像生成部16dを備えている。なお、インター予測画像生成部16は、単方向予測、及び、双方向予測を行うことが可能であるが、以下では、双方向予測(重み付き予測)を行うための構成について説明する。
次に、インター予測画像生成部16によるインター予測画像Pred_Interの生成処理について、図4を参照して説明する。
同様に、参照値算出部16aは、フレーム間距離tdを、参照画像L0の表示順序POC_L0と、参照画像L1の表示順序POC_L1とを用いて、以下の式(2)より算出する。
なお、フレーム間距離tb、及び、フレーム間距離tdは、以下の式(3)、及び、式(4)を用いて算出してもよい。
td=abs(POC_cur-POC_L1)
+abs(POC_cur-POC_L0) (4)
ここで、abs(・・・)は、括弧内の値の絶対値を取ることを表している。すなわち、式(1)、及び、式(2)によって表されるフレーム間距離tb、及び、フレーム間距離tdは、何れも、負の値もとり得るが、式(3)、及び、(4)によって表される間距離tb、及び、フレーム間距離tdは、正の値のみをとる。
mc2(x,y)=L1(x+mv2x,y+mv2y) (6)
ここで、(x、y)は、動き補償画像mc1及び動き補償画像mc2における対象画素の座標を表し、(mv1x,mv1y)は、動きベクトルmv1のx成分、及び、y成分を表しており、(mv2x,mv2y)は、動きベクトルmv2のx成分、及び、y成分を表している。また、L0(x、y)は、参照画像L0の座標(x、y)における画素値を表し、L1(x、y)は、参照画像L1の座標(x、y)における画素値を表している。
ここで、oはオフセット値を示し、(x、y)は、対象画素の座標を表している。また、重み係数w1及びw2は、w1+w2=1を満たす重み係数を表している。
以下では、重み係数算出部16bによる、重み係数w1及びw2の算出処理の具体例について説明する。
以下では、重み係数算出部16bによる、重み係数w1及びw2の算出処理の第1の例について説明する。
w1=1-w2 (9)
ここで、clip3(引数1,引数2,引数3)は、引数3が引数1より小さい値である場合(引数1>引数3)、引数1をとり、引数3が引数2より大きい値である場合(引数2<引数3)、引数2をとり、引数3が引数1以上引数2以下の値である場合(引数1≦引数3≦引数2)、引数3をとる関数を表している。すなわち、clip3(引数1,引数2,引数3)は、異なる2つの閾値(引数1および引数2)を用いて値の範囲を限定するクリップ処理を表している。
w2=clip3(th1×64,th2×64,w2a) (8’)
w1=64-w2 (9’)
ここで、w2aは、w2を算出するために用いられる一時的な変数である。
w2=clip3(21,43,tb/td)
と表される。この場合、予め与えられた整数を乗ずることによって得られる値が、少数であった場合、小数点以下を四捨五入することによって、整数表現に丸めるため、クリップ処理の引数1、及び、引数2の値は、整数になる。
pred(x,y)=(w1×mc1(x,y)+w2×mc2(x,y)+o)/64 (7’)
を用いることにより生成される。
以下では、重み係数算出部16bによる、重み係数w1及びw2の算出処理の第2の例について説明する。
w2b={(tb/td-0.5)×r+0.5} (10)
w2=clip3(th1,th2,w2b) (11)
ここで、w2bは、w2を算出するために一時的に用いる変数であり、式(10)から明らかなように、w2bの値は、与えられたtb/tdの値よりも0.5に近い値をとる。また、式(10)におけるrは、tb/tdの値に比べて、w2bの値をどの程度0.5に近づけるかを規定する定数である。また、th1及びth2は、具体例1-1と同様に、クリップ処理において用いられる閾値である。r、th1及びth2の具体的な値は、符号化効率が最大となるように予め設定されている。本例では、r=3/4、th1=0.33、および、th2=0.67とするが、0<r<1、かつ、th1<0.5<th2を満たせば、これに限定されるものではない。
以下では、重み係数算出部16bによる、重み係数w1及びw2の算出処理の第3の例について説明する。
ここで、th1’、及び、th2’は、パーティションサイズに応じて変化するクリップ処理の下限閾値及び上限閾値である。例えば、パーティションサイズが16×16以上である場合、(th1’,th2’)を(0.30,0.70)に設定し、パーティションサイズが16×8もしくは8×16である場合、(th1’,th2’)を(0.35,0.65)に設定し、パーティションサイズが8×8である場合、(th1’,th2’)を(0.40,0.60)に設定し、パーティションサイズが8×4もしくは4×8である場合、(th1’,th2’)を(0.45,0.55)に設定し、パーティションサイズが4×4である場合、(th1’,th2’)を(0.50,0.50)に設定する。
以下では、重み係数算出部16bによる、重み係数w1及びw2の算出処理の第4の例について説明する。
w2=(tb/td)×r1+0.5×r2 (13)
ここで、r1及びr2は、最適重み係数を近似する分母が共通する有理数係数である。パーティションサイズ毎に最小二乗法により決定した最適重み係数を図5に示す。各パーティションサイズに応じた係数r1及びr2は、図5に示すグラフに基づいて、例えば以下のように設定することができる。
以下では、重み係数算出部16bによる、重み係数w1及びw2の算出処理の第5の例について説明する。
pred=w1×mc1+(1‐w1)×mc2
と表す。ここで、mc1、及び、mc2は、動き補償画像の画素値を表す。
pred=w1×mc1+(1-w1)×mc2
と表す。ここで、mc1及びmc2は、動き補償画像の画素値を表す。時間変化による誤差n1、及び、n2を、一定時間で発生する誤差ncを用いて、n1=nc×t1、及び、n2=nc×t2と表す。ここで、t1及びt2は予測対象画像(復号中の復号画像)からの時間差を表す。図6に参照フレームと予測対象フレームとの位置関係と時間tの関係の一例を示す。図6より、予測対象画像を起点(t=0)とした場合、t1は、予測対象画像より前の時間(t1<0)を表し、t2は、予測対象画像より後の時間(t2>0)を表す。
E=|orig-pred|2
=|w1×n1+(1-w1)×n2|2
ここで、n1=nc1×t1、及び、n2=nc2×t2を代入すると、
E=|w1×nc1×t1+(1-w1)×nc2×t2|2
=w12×NC1×t12+(1-w1)2×NC2×t22
を得る。ここで、NC1およびNC2は、NC1=nc12、および、NC2=nc22によって与えられる。
dE/dw1=2×w1×NC1×t12-2×(1-w1)×NC2×t22
=2×(NC1×t12+NC2×t22)×w1-2×NC2×t22
上記式より、誤差Eを最小化するw1は、上記(dE/dw1)=0を解いて、以下の式(17’)のように表される。
w1=(NC2×t22)/(NC1×t12+NC2×t22) (17’)
また、重み係数w2は、式(9)を用いて、以下の式(18’)のように表される。
w2=(NC1×t12)/(NC1×t12+NC2×t22) (18’)
また、nc1とnc2とが互いにほぼ等しい場合、すなわち、NC1とNC2とが互いにほぼ等しい場合には、w1およびw2は、以下の式(17’’)および(18’’)のように表される。
w1=t22/(t12+t22) (17’’)
w2=t12/(t12+t22) (18’’)
重み係数算出部16bは、このように、式(17’)、式(18’)、式(17’’)、及び、式(18’’)によって重み係数w1及びw2を算出する構成としてもよい。
本実施形態に係る動画像符号化装置2の構成について図7~図9を参照して説明する。動画像符号化装置2は、その一部に、H.264/MPEG-4.AVC、及び、KTAソフトウェアに採用されている技術を含む復号装置である。
図8は、インター予測画像生成部26の構成を示すブロック図である。図8に示すように、インター予測画像生成部26は、参照値算出部26a、重み係数算出部26b、動き補償部26c、及び、予測画像生成部26dを備えている。
以下では、重み係数算出部26bによる、重み係数w1及びw2の算出処理の具体例について説明する。重み係数算出部26bは、以下の(具体例1-1’)~(具体例1-5’)に挙げる算出処理のうち、何れかの算出処理を行う。
以下では、重み係数算出部26bによる、重み係数w1及びw2の算出処理の第1の例について説明する。本例では、動画像復号装置1の重み係数算出部16bによる、重み係数w1及びw2の算出処理の(具体例1-1)と同様に式(8)及び式(9)を用いて重み係数w1及びw2を算出する。
以下では、重み係数算出部26bによる、重み係数w1及びw2の算出処理の第2の例について説明する。本例では、動画像復号装置1の重み係数算出部16bによる、重み係数w1及びw2の算出処理の(具体例1-2)と同様に式(10)、式(11)、及び、式(9)を用いて重み係数w1及びw2を算出する。
以下では、重み係数算出部26bによる、重み係数w1及びw2の算出処理の第3の例について説明する。本例では、動画像復号装置1の重み係数算出部16bによる、重み係数w1及びw2の算出処理の(具体例1-3)と同様に式(12)及び式(9)を用いて重み係数w1及びw2を算出する。
以下では、重み係数算出部26bによる、重み係数w1及びw2の算出処理の第4の例について説明する。本例では、動画像復号装置1の重み係数算出部16bによる、重み係数w1及びw2の算出処理の(具体例1-4)と同様に式(13)及び式(9)を用いて重み係数w1及びw2を算出する。
以下では、重み係数算出部26bによる、重み係数w1及びw2の算出処理の第5の例について説明する。本例では、動画像復号装置1の重み係数算出部16bによる、重み係数w1及びw2の算出処理の(具体例1-5)と同様に式(17)及び式(18)を用いて重み係数w1及びw2を算出する。
(動画像復号装置3)
本実施形態に係る動画像復号装置3の構成について図10~図12を参照して説明する。図10は、動画像復号装置3の構成を示すブロック図である。なお、図10では、図1において示した各ブロックと同一のブロックに関しては、同一の参照符号を付し、その説明を省略する。
次に、インター予測画像生成部16’の構成について、図11を参照して説明する。図11は、インター予測画像生成部16’の構成を示すブロック図である。図11に示すように、インター予測画像生成部16’は、参照値算出部16a’、重み係数算出部16b’、動き補償部16c、及び、予測画像生成部16dを備えている。なお、インター予測画像生成部16’は、単方向予測、及び、双方向予測を行うことが可能であるが、以下では、双方向予測(重み付き予測)を行うための構成について説明する。
次に、インター予測画像生成部16’によるインター予測画像Pred_Interの生成処理について、図12を参照して説明する。図12は、参照画像L0を参照した動き補償に用いられる動きベクトルmv1と参照画像L1を参照した動き補償に用いられる動きベクトルmv2の一例を表した図である。
ここで、oはオフセット値を示し、(x、y)は、対象画素の座標を表している。また、重み係数w1’及びw2’は、重み係数算出部16b’によって算出された、w1’+w2’=1を満たす重み係数を表している。
以下では、重み係数算出部16b’による、重み係数w1’及びw2’の算出処理の具体例について説明する。
以下では、重み係数算出部16b’による、重み係数w1’及びw2’の算出処理の第1の例について説明する。
w2’=clip3(th1,th2,ml1/(ml1+ml2)) (21)
w1’=1-w2’ (22)
ここで、th1、及び、th2は、クリップ処理における下限閾値、及び、上限閾値を表しており、符号化効率の高い最適な重み係数w1’及びw2’を算出できる値に設定する。本例では、th1=0.33、th2=0.67とするが、th1<0.5<th2を満たせば、これに限定されるものではない。
以下では、重み係数算出部16b’による、重み係数w1’及びw2’の算出処理の第2の例について説明する。
以下では、重み係数算出部16b’による、重み係数w1’及びw2’の算出処理の第3の例について説明する。
w2’=ml1/(ml1+ml2) (23)
また、重み係数算出部16b’は、tb/td=0.5以外の場合、重み係数w2’をw2’=tb/tdに設定する。また、第1の実施形態で計算される値に設定することも好適である。
本実施形態に係る動画像符号化装置4の構成について図13~図14を参照して説明する。図13は、動画像符号化装置4の構成を示すブロック図である。なお、図13では、図7において示した各ブロックと同一のブロックに関しては、同一の参照符号を付し、その説明を省略する。
図14は、インター予測画像生成部26’の構成を示すブロック図である。図14に示すように、インター予測画像生成部26’は、参照値算出部26a’、重み係数算出部26b’、動き補償部26c、及び、予測画像生成部26dを備えている。
以下では、重み係数算出部26b’による、重み係数w1’及びw2’の算出処理の具体例について説明する。重み係数算出部26b’は、以下の(具体例2-1’)~(具体例2-3’)に挙げる算出処理のうち、何れかの算出処理を行う。
以下では、重み係数算出部26b’による、重み係数w1’及びw2’の算出処理の第1の例について説明する。本例では、動画像復号装置3の重み係数算出部16b’による、重み係数w1’及びw2’の算出処理の(具体例2-1)と同様に式(21)、及び、式(22)を用いて重み係数w1’及びw2’を算出する。
以下では、重み係数算出部26b’による、重み係数w1’及びw2’の算出処理の第2の例について説明する。本例では、動画像復号装置3の重み係数算出部16b’による、重み係数w1’及びw2’の算出処理の(具体例2-2)と同様に、重み係数w2’は、動きベクトル長ml1及びml2を用いて、(具体例2-2)で説明した(1)から(3)の何れかの値に設定する。
以下では、重み係数算出部26b’による、重み係数w1’及びw2’の算出処理の第3の例について説明する。本例では、動画像復号装置3の重み係数算出部16b’による、重み係数w1’及びw2’の算出処理の(具体例2-3)と同様に式(23)を用いるか、もしくは、w2’=0.5に重み係数w2’を設定する。
(動画像復号装置5)
本実施形態に係る動画像復号装置5の構成について図15~図17を参照して説明する。図15は、動画像復号装置5の構成を示すブロック図である。なお、図15では、図1において示した各ブロックと同一のブロックに関しては、同一の参照符号を付し、その説明を省略する。
次に、インター予測画像生成部16”の構成について、図16を参照して説明する。図16は、インター予測画像生成部16”の構成を示すブロック図である。図16に示すように、インター予測画像生成部16”は、参照値算出部16a”、重み係数算出部16b”、動き補償部16c、及び、予測画像生成部16dを備えている。なお、インター予測画像生成部16”は、単方向予測、及び、双方向予測を行うことが可能であるが、以下では、双方向予測(重み付き予測)を行うための構成について説明する。
次に、インター予測画像生成部16”によるインター予測画像Pred_Interの生成処理について、図17を参照して説明する。図17は、参照画像L0を参照した動き補償に用いられる動きベクトルmv1と参照画像L1を参照した動き補償に用いられる動きベクトルmv2の一例を表した図である。
ここで、oはオフセット値を示し、(x、y)は、対象画素の座標を表している。また、重み係数w1”及びw2”は、重み係数算出部16b”によって算出された、w1”+w2”=1を満たす重み係数を表している。
以下では、重み係数算出部16b”による、重み係数w1”及びw2”の算出処理の具体例について説明する。
以下では、重み係数算出部16b”による、重み係数w1”及びw2”の算出処理の第1の例について説明する。
w2”=clip3(th1,th2,QP1/(QP1+QP2)) (26)
w1”=1-w2” (27)
ここで、th1、及び、th2は、クリップ処理における下限閾値、及び、上限閾値を表しており、符号化効率の高い最適な重み係数w1”及びw2”を算出できる値に設定する。本例では、th1=0.33、th2=0.67とするが、th1<0.5<th2を満たせば、これに限定されるものではない。
以下では、重み係数算出部16b”による、重み係数w1”及びw2”の算出処理の第2の例について説明する。
以下では、重み係数算出部16b”による、重み係数w1”及びw2”の算出処理の第3の例について説明する。
w2”=QP1/(QP1+QP2) (28)
また、重み係数算出部16b’は、tb/td=0.5以外の場合、重み係数w2”=tb/tdに設定する。また、第1の実施形態で計算される値とすることも好適である。
本実施形態に係る動画像符号化装置6の構成について図18~図19を参照して説明する。図18は、動画像符号化装置6の構成を示すブロック図である。なお、図18では、図7において示した各パーティションと同一のパーティションに関しては、同一の参照符号を付し、その説明を省略する。
図19は、インター予測画像生成部26”の構成を示すブロック図である。図19に示すように、インター予測画像生成部26”は、参照値算出部26a”、重み係数算出部26b”、動き補償部26c、及び、予測画像生成部26dを備えている。
以下では、重み係数算出部26b”による、重み係数w1”及びw2”の算出処理の具体例について説明する。重み係数算出部26b”は、以下の(具体例3-1’)~(具体例3-3’)に挙げる算出処理のうち、何れかの算出処理を行う。
以下では、重み係数算出部26b”による、重み係数w1”及びw2”の算出処理の第1の例について説明する。本例では、動画像復号装置3の重み係数算出部16b’による、重み係数w1”及びw2”の算出処理の(具体例3-1)と同様に式(26)及び式(27)を用いて重み係数w1”及びw2”を算出する。
以下では、重み係数算出部26b”による、重み係数w1”及びw2”の算出処理の第2の例について説明する。本例では、動画像復号装置5の重み係数算出部16b”による、重み係数w1”及びw2”の算出処理の(具体例3-2)と同様に、重み係数w2は、量子化ステップQP1及びQP2を用いて、(具体例3-2)で説明した(1)から(3)の何れかの値に設定する。
以下では、重み係数算出部26b”による、重み係数w1”及びw2”の算出処理の第3の例について説明する。本例では、動画像復号装置5の重み係数算出部16b”による、重み係数w1”及びw2”の算出処理の(具体例3-3)と同様に式(28)を用いるか、もしくは、w2”=0.5に重み係数w2”設定する。
最後に、上述した動画像復号装置1、3、5及び動画像符号化装置2、4、6の各ブロックは、集積回路(ICチップ)上に形成された論理回路によってハードウェア的に実現してもよいし、CPU(Central Processing Unit)を用いてソフトウェア的に実現してもよい。
以上のように、本発明に係る予測画像生成装置は、復号画像P1及びP2の各々を参照画像とし、復号画像Pに対する動き補償画像mc1及びmc2を生成する動き補償画像生成手段と、上記動き補償画像生成手段によって生成された動き補償画像mc1及びmc2の各々に乗じる重み係数w1及びw2の値を設定する重み係数設定手段と、上記重み係数設定手段によって設定された重み係数w1及びw2を用いて、上記動き補償画像生成手段によって生成された動き補償画像mc1及びmc2の加重平均w1×mc1+w2×mc2にオフセット値を加算することによって、復号画像Pに対する予測画像Predを生成する予測画像生成手段と、を備えた予測画像生成装置である。
16 インター予測画像生成部(予測画像生成装置)
16a 参照値算出部
16b 重み係数算出部(重み係数設定手段)
16c 動き補償部(動き補償画像生成手段)
16d 予測画像生成部(予測画像生成手段)
2 動画像符号化装置
26 インター予測画像生成部(予測画像生成装置)
26a 参照値算出部
26b 重み係数算出部(重み係数設定手段)
26c 動き補償部(動き補償画像生成手段)
26d 予測画像生成部(予測画像生成手段)
27 動きベクトル検出部
Claims (18)
- 復号画像P1及びP2の各々を参照画像とし、復号画像Pに対する動き補償画像mc1及びmc2を生成する動き補償画像生成手段と、
上記動き補償画像生成手段によって生成された動き補償画像mc1及びmc2の各々に乗じる重み係数w1及びw2の値を設定する重み係数設定手段と、
上記重み係数設定手段によって設定された重み係数w1及びw2を用いて、上記動き補償画像生成手段によって生成された動き補償画像mc1及びmc2の加重平均w1×mc1+w2×mc2にオフセット値を加算することによって、復号画像Pに対する予測画像Predを生成する予測画像生成手段と、を備え、
上記重み係数設定手段は、復号画像P1と復号画像P2とのフレーム間距離をtd、復号画像P1と復号画像Pとの距離をtbとして、重み係数w2の値を、0<|w2-0.5|<|tb/td-0.5|となるように設定する、ことを特徴とする予測画像生成装置。 - 上記重み係数設定手段は、重み係数w2の値を、閾値Th1及びTh2(Th1<0.5<Th2)を用いてclip3(Th1,Th2,tb/td)に設定する、
ことを特徴とする請求項1に記載の予測画像生成装置。 - 上記重み係数設定手段は、重み係数w2の値を、係数α(0<α<1)を用いて(tb/td-0.5)α+0.5に設定するか、又は、係数α(0<α<1)並びに閾値Th1及びTh2(Th1<0.5<Th2)を用いてclip3(Th1,Th2,(tb/td-0.5)α+0.5)に設定する、
ことを特徴とする請求項1に記載の予測画像生成装置。 - 上記重み係数設定手段は、予測画像Predを生成すべきパーティションのサイズに応じて上記係数αを設定する、
ことを特徴とする請求項3に記載の予測画像生成装置。 - 上記重み係数設定手段は、予測画像Predを生成すべきパーティションのサイズに応じて上記閾値Th1及びTh2を設定する、
ことを特徴とする請求項2から4までの何れか1項に記載の予測画像生成装置。 - 復号画像P1及びP2の各々を参照画像とし、復号画像Pに対する動き補償画像mc1及びmc2を生成する動き補償画像生成手段と、
上記動き補償画像生成手段によって生成された動き補償画像mc1及びmc2の各々に乗じる重み係数w1及びw2の値を設定する重み係数設定手段と、
上記重み係数設定手段によって設定された重み係数w1及びw2を用いて、上記動き補償画像生成手段によって生成された動き補償画像mc1及びmc2の加重平均w1×mc1+w2×mc2にオフセット値を加算することによって、復号画像Pに対する予測画像Predを生成する予測画像生成手段と、を備え、
上記重み係数設定手段は、動き補償画像mc1及びmc2を得るための動き補償に用いた動きベクトルmv1及びmv2の大きさ|mv1|及び|mv2|が|mv1|<|mv2|であるとき、重み係数w1及びw2の値を|w1|≧|w2|となるように設定する、ことを特徴とする予測画像生成装置。 - 上記重み係数設定手段は、重み係数w2の値を、|mv1|/(|mv1|+|mv2|)に設定するか、又は、閾値Th1及びTh2(Th1<0.5<Th2)を用いてclip3(Th1,Th2,|mv1|/(|mv1|+|mv2|))に設定する、
ことを特徴とする請求項6に記載の予測画像生成装置。 - 上記重み係数設定手段は、kを正の定数として、重み係数w2の値を、(1)|mv1|<|mv2|-kのとき0.5よりも小さい値に設定し、(2)|mv2|-k≦|mv1|≦|mv2|+kのとき0.5に設定し、(3)|mv2|+k<|mv1|のとき0.5よりも大きい値に設定する、
ことを特徴とする請求項6に記載の予測画像生成装置。 - 上記重み係数設定手段は、復号画像P1と復号画像Pとのフレーム間距離が復号画像P2と復号画像Pとのフレーム間距離に等しいときに限って、重み係数w1及びw2の値を動きベクトルmv1及びmv2の大きさ|mv1|及び|mv2|に応じて設定する、
ことを特徴とする請求項6から8までの何れか1項に記載の予測画像生成装置。 - 復号画像P1及びP2の各々を参照画像とし、復号画像Pに対する動き補償画像mc1及びmc2を生成する動き補償画像生成手段と、
上記動き補償画像生成手段によって生成された動き補償画像mc1及びmc2の各々に乗じる重み係数w1及びw2の値を設定する重み係数設定手段と、
上記重み係数設定手段によって設定された重み係数w1及びw2を用いて、上記動き補償画像生成手段によって生成された動き補償画像mc1及びmc2の加重平均w1×mc1+w2×mc2にオフセット値を加算することによって、復号画像Pに対する予測画像Predを生成する予測画像生成手段と、を備え、
上記重み係数設定手段は、復号画像P1及びP2の各々を得るための逆量子化に用いた量子化ステップQP1及びQP2がQP1<QP2であるとき、上記重み係数w1及びw2の値を|w1|≧|w2|となるように設定する、ことを特徴とする予測画像生成装置。 - 上記重み係数設定手段は、重み係数w2の値を、QP1/(QP1+QP2)に設定するか、又は、閾値Th1及びTh2(Th1<0.5<Th2)を用いてclip3(Th1,Th2,QP1/(QP1+QP2))に設定する、
ことを特徴とする請求項10に記載の予測画像生成装置。 - 上記重み係数設定手段は、kを正の定数として、重み係数w2の値を、(1)QP1<QP2-kのとき0.5よりも小さい値に設定し、(2)QP2-k≦QP1≦QP2+kのとき0.5に設定し、(3)QP2+k<QP1のとき0.5よりも大きい値に設定する、
ことを特徴とする請求項10に記載の予測画像生成装置。 - 上記重み係数設定手段は、復号画像P1と復号画像Pとのフレーム間距離が復号画像P2と復号画像Pとのフレーム間距離に等しいときに限って、重み係数w1及びw2の値を量子化ステップQP1及びQP2に応じて設定する、
ことを特徴とする請求項10から12までの何れか1項に記載の予測画像生成装置。 - 請求項1から13までの何れか1項に記載の予測画像生成装置を備えた動画像復号装置であって、
上記予測画像生成装置が他の復号画像P1及びP2を参照して生成した予測画像Predに符号化データから復号した予測残差を加算することによって、復号画像Pを生成する、ことを特徴とする動画像復号装置。 - 請求項1から13までの何れか1項に記載の予測画像生成装置を備えた動画像符号化装置であって、
上記予測画像生成装置が局所復号画像P1及びP2を参照して生成した予測画像Predを符号化対象画像から減算して得た予測残差を符号化することによって、符号化データを生成する
ことを特徴とする動画像符号化装置。 - 復号画像P1及びP2の各々を参照画像とし、復号画像Pに対する動き補償画像mc1及びmc2を生成する動き補償画像生成工程と、
上記動き補償画像生成工程にて生成された動き補償画像mc1及びmc2の各々に乗じる重み係数w1及びw2の値を設定する重み係数設定工程と、
上記重み係数設定工程にて設定された重み係数w1及びw2を用いて、上記動き補償画像生成工程にて生成された動き補償画像mc1及びmc2の加重平均w1×mc1+w2×mc2にオフセット値を加算することによって、復号画像Pに対する予測画像Predを生成する予測画像生成工程と、を含み、
上記重み係数設定工程において、復号画像P1と復号画像P2とのフレーム間距離をtd、復号画像P1と復号画像Pとの距離をtbとして、重み係数w2の値を、0<|w2-0.5|<|tb/td-0.5|となるように設定する、ことを特徴とする予測画像生成方法。 - 復号画像P1及びP2の各々を参照画像とし、復号画像Pに対する動き補償画像mc1及びmc2を生成する動き補償画像生成工程と、
上記動き補償画像生成工程にて生成された動き補償画像mc1及びmc2の各々に乗じる重み係数w1及びw2の値を設定する重み係数設定工程と、
上記重み係数設定工程にて設定された重み係数w1及びw2を用いて、上記動き補償画像生成工程にて生成された動き補償画像mc1及びmc2の加重平均w1×mc1+w2×mc2にオフセット値を加算することによって、復号画像Pに対する予測画像Predを生成する予測画像生成工程と、を含み、
上記重み係数設定工程において、動き補償画像mc1及びmc2を得るための動き補償に用いた動きベクトルmv1及びmv2の大きさ|mv1|及び|mv2|が|mv1|<|mv2|であるとき、重み係数w1及びw2の値を|w1|≧|w2|となるように設定する、ことを特徴とする予測画像生成方法。 - 復号画像P1及びP2の各々を参照画像とし、復号画像Pに対する動き補償画像mc1及びmc2を生成する動き補償画像生成工程と、
上記動き補償画像生成工程にて生成された動き補償画像mc1及びmc2の各々に乗じる重み係数w1及びw2の値を設定する重み係数設定工程と、
上記重み係数設定工程にて設定された重み係数w1及びw2を用いて、上記動き補償画像生成工程にて生成された動き補償画像mc1及びmc2の加重平均w1×mc1+w2×mc2にオフセット値を加算することによって、復号画像Pに対する予測画像Predを生成する予測画像生成工程と、を含み、
上記重み係数設定工程において、復号画像P1及びP2の各々を得るための逆量子化に用いた量子化ステップQP1及びQP2がQP1<QP2であるとき、上記重み係数w1及びw2の値を|w1|≧|w2|となるように設定する、ことを特徴とする予測画像生成方法。
Priority Applications (4)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
US13/697,261 US10034012B2 (en) | 2010-05-13 | 2011-04-28 | Prediction image generation device, moving image decoding device, and moving image coding device |
CN201180033380.3A CN102972027B (zh) | 2010-05-13 | 2011-04-28 | 预测图像生成装置、动图像解码装置、动图像编码装置以及预测图像生成方法 |
CA2799144A CA2799144C (en) | 2010-05-13 | 2011-04-28 | Prediction image generation device, moving image decoding device, and moving image coding device |
JP2012514767A JP5784596B2 (ja) | 2010-05-13 | 2011-04-28 | 予測画像生成装置、動画像復号装置、及び、動画像符号化装置 |
Applications Claiming Priority (2)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
JP2010-111476 | 2010-05-13 | ||
JP2010111476 | 2010-05-13 |
Publications (1)
Publication Number | Publication Date |
---|---|
WO2011142277A1 true WO2011142277A1 (ja) | 2011-11-17 |
Family
ID=44914333
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
PCT/JP2011/060428 WO2011142277A1 (ja) | 2010-05-13 | 2011-04-28 | 予測画像生成装置、動画像復号装置、及び、動画像符号化装置 |
Country Status (5)
Country | Link |
---|---|
US (1) | US10034012B2 (ja) |
JP (1) | JP5784596B2 (ja) |
CN (3) | CN107071410A (ja) |
CA (1) | CA2799144C (ja) |
WO (1) | WO2011142277A1 (ja) |
Cited By (1)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
JP2020522185A (ja) * | 2017-08-14 | 2020-07-27 | グーグル エルエルシー | 複合動き補償予測 |
Families Citing this family (10)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US9888817B2 (en) | 2014-12-17 | 2018-02-13 | Omachron Intellectual Property Inc. | Surface cleaning apparatus |
US9438906B2 (en) | 2011-03-03 | 2016-09-06 | Sun Patent Trust | Method of encoding an image into a coded image, method of decoding a coded image, and apparatuses thereof |
KR20160065939A (ko) * | 2013-11-05 | 2016-06-09 | 애리스 인터프라이지즈, 인크. | 고정밀도 데이터에 대한 비트 심도 변수를 이용한 가중화된 예측 신택스 및 시맨틱들의 간략화된 프로세싱 |
TWI536811B (zh) * | 2013-12-27 | 2016-06-01 | 財團法人工業技術研究院 | 影像處理方法與系統、解碼方法、編碼器與解碼器 |
WO2019004283A1 (ja) * | 2017-06-28 | 2019-01-03 | シャープ株式会社 | 動画像符号化装置及び動画像復号装置 |
KR101997681B1 (ko) * | 2018-06-11 | 2019-07-08 | 광운대학교 산학협력단 | 양자화 파라미터 기반의 잔차 블록 부호화/복호화 방법 및 장치 |
US11006799B2 (en) | 2018-08-13 | 2021-05-18 | Omachron Intellectual Property Inc. | Cyclonic air treatment member and surface cleaning apparatus including the same |
US11192122B2 (en) | 2018-08-13 | 2021-12-07 | Omachron Intellectual Property Inc. | Cyclonic air treatment member and surface cleaning apparatus including the same |
US11013384B2 (en) | 2018-08-13 | 2021-05-25 | Omachron Intellectual Property Inc. | Cyclonic air treatment member and surface cleaning apparatus including the same |
WO2020059877A1 (ja) * | 2018-09-21 | 2020-03-26 | シャープ株式会社 | 画像復号装置および画像符号化装置 |
Citations (4)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
JP2004194285A (ja) * | 2002-11-20 | 2004-07-08 | Matsushita Electric Ind Co Ltd | 動画像予測方法、動画像符号化方法及び装置、動画像復号化方法及び装置 |
JP2004242276A (ja) * | 2002-08-08 | 2004-08-26 | Matsushita Electric Ind Co Ltd | 動画像の符号化方法および復号化方法 |
JP2006101322A (ja) * | 2004-09-30 | 2006-04-13 | Toshiba Corp | 情報処理装置および同装置で用いられるプログラム |
JP2007081518A (ja) * | 2005-09-12 | 2007-03-29 | Victor Co Of Japan Ltd | 動画像符号化装置および動画像符号化方法 |
Family Cites Families (19)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US5412435A (en) * | 1992-07-03 | 1995-05-02 | Kokusai Denshin Denwa Kabushiki Kaisha | Interlaced video signal motion compensation prediction system |
US6404813B1 (en) * | 1997-03-27 | 2002-06-11 | At&T Corp. | Bidirectionally predicted pictures or video object planes for efficient and flexible video coding |
EP1045593A4 (en) * | 1998-11-04 | 2007-06-06 | Mitsubishi Electric Corp | DECODER AND IMAGE ENCODER |
US7450641B2 (en) * | 2001-09-14 | 2008-11-11 | Sharp Laboratories Of America, Inc. | Adaptive filtering based upon boundary strength |
EP1827029A1 (en) * | 2002-01-18 | 2007-08-29 | Kabushiki Kaisha Toshiba | Video decoding method and apparatus |
ES2650068T3 (es) | 2002-08-08 | 2018-01-16 | Godo Kaisha Ip Bridge 1 | Procedimiento de codificación y procedimiento de descodificación de imágenes en movimiento |
CA2473767C (en) | 2002-11-20 | 2012-10-23 | Matsushita Electric Industrial Co., Ltd. | Moving picture prediction method, moving picture coding method and apparatus, and moving picture decoding method and apparatus |
EP1636998A2 (en) * | 2003-06-25 | 2006-03-22 | Thomson Licensing | Method and apparatus for weighted prediction estimation using a displaced frame differential |
US7953152B1 (en) * | 2004-06-28 | 2011-05-31 | Google Inc. | Video compression and encoding method |
DE602005015067D1 (de) | 2004-09-30 | 2009-08-06 | Toshiba Kk | Informationsbearbeitungsgerät und Programm |
JP2006101321A (ja) * | 2004-09-30 | 2006-04-13 | Toshiba Corp | 情報処理装置および同装置で用いられるプログラム |
US7663701B2 (en) * | 2005-04-11 | 2010-02-16 | Ati Technologies, Inc. | Systems, methods, and apparatus for noise reduction |
WO2008140656A2 (en) * | 2007-04-03 | 2008-11-20 | Gary Demos | Flowfield motion compensation for video compression |
JP4780053B2 (ja) * | 2007-07-23 | 2011-09-28 | ソニー株式会社 | 画像処理装置、画像処理方法、及び、プログラム |
WO2009041215A1 (ja) * | 2007-09-25 | 2009-04-02 | Sharp Kabushiki Kaisha | 動画像符号化装置及び動画像復号装置 |
EP2061248A1 (en) * | 2007-11-13 | 2009-05-20 | IBBT vzw | Motion estimation and compensation process and device |
US8750378B2 (en) * | 2008-09-23 | 2014-06-10 | Qualcomm Incorporated | Offset calculation in switched interpolation filters |
JP5325638B2 (ja) * | 2008-11-26 | 2013-10-23 | 日立コンシューマエレクトロニクス株式会社 | 画像復号化方法 |
US9161058B2 (en) * | 2010-03-27 | 2015-10-13 | Texas Instruments Incorporated | Method and system for detecting global brightness change for weighted prediction in video encoding |
-
2011
- 2011-04-28 WO PCT/JP2011/060428 patent/WO2011142277A1/ja active Application Filing
- 2011-04-28 CN CN201611048718.3A patent/CN107071410A/zh active Pending
- 2011-04-28 US US13/697,261 patent/US10034012B2/en not_active Expired - Fee Related
- 2011-04-28 CA CA2799144A patent/CA2799144C/en not_active Expired - Fee Related
- 2011-04-28 CN CN201180033380.3A patent/CN102972027B/zh not_active Expired - Fee Related
- 2011-04-28 CN CN201611042219.3A patent/CN107071409A/zh active Pending
- 2011-04-28 JP JP2012514767A patent/JP5784596B2/ja not_active Expired - Fee Related
Patent Citations (4)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
JP2004242276A (ja) * | 2002-08-08 | 2004-08-26 | Matsushita Electric Ind Co Ltd | 動画像の符号化方法および復号化方法 |
JP2004194285A (ja) * | 2002-11-20 | 2004-07-08 | Matsushita Electric Ind Co Ltd | 動画像予測方法、動画像符号化方法及び装置、動画像復号化方法及び装置 |
JP2006101322A (ja) * | 2004-09-30 | 2006-04-13 | Toshiba Corp | 情報処理装置および同装置で用いられるプログラム |
JP2007081518A (ja) * | 2005-09-12 | 2007-03-29 | Victor Co Of Japan Ltd | 動画像符号化装置および動画像符号化方法 |
Cited By (2)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
JP2020522185A (ja) * | 2017-08-14 | 2020-07-27 | グーグル エルエルシー | 複合動き補償予測 |
JP2022069546A (ja) * | 2017-08-14 | 2022-05-11 | グーグル エルエルシー | 複合動き補償予測 |
Also Published As
Publication number | Publication date |
---|---|
US10034012B2 (en) | 2018-07-24 |
CN102972027A (zh) | 2013-03-13 |
JPWO2011142277A1 (ja) | 2013-07-22 |
CA2799144A1 (en) | 2011-11-17 |
JP5784596B2 (ja) | 2015-09-24 |
CN107071410A (zh) | 2017-08-18 |
CA2799144C (en) | 2016-03-15 |
CN102972027B (zh) | 2016-12-21 |
CN107071409A (zh) | 2017-08-18 |
US20130058417A1 (en) | 2013-03-07 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
JP5784596B2 (ja) | 予測画像生成装置、動画像復号装置、及び、動画像符号化装置 | |
US11843783B2 (en) | Predictive motion vector coding | |
JP5261774B2 (ja) | 動画像スケーラブル符号化方法及び復号方法、それらの装置、それらのプログラム並びにプログラムを記録した記録媒体 | |
JP5061179B2 (ja) | 照明変化補償動き予測符号化および復号化方法とその装置 | |
JP5185266B2 (ja) | 動画像スケーラブル符号化方法及び復号方法、それらの装置、それらのプログラム並びにプログラムを記録した記録媒体 | |
JP5277257B2 (ja) | 動画像復号化方法および動画像符号化方法 | |
JP4752631B2 (ja) | 画像符号化装置、及び画像符号化方法 | |
JP2005191706A (ja) | 動画像符号化方法及びその装置 | |
WO2015010319A1 (zh) | 一种基于p帧的多假设运动补偿编码方法 | |
US20140133546A1 (en) | Video encoding device, video decoding device, video encoding method, video decoding method, video encoding program, and video decoding program | |
JP5590269B1 (ja) | 映像符号化方法、映像符号化装置および映像符号化プログラム | |
JP2019193182A (ja) | 符号化装置、符号化方法、及び符号化プログラム | |
JP2022070176A (ja) | 符号化装置及びプログラム | |
JP2007201675A (ja) | 動画像符号化装置および動画像符号化方法 |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
WWE | Wipo information: entry into national phase |
Ref document number: 201180033380.3 Country of ref document: CN |
|
121 | Ep: the epo has been informed by wipo that ep was designated in this application |
Ref document number: 11780534 Country of ref document: EP Kind code of ref document: A1 |
|
ENP | Entry into the national phase |
Ref document number: 2799144 Country of ref document: CA |
|
WWE | Wipo information: entry into national phase |
Ref document number: 2012514767 Country of ref document: JP Ref document number: 13697261 Country of ref document: US |
|
NENP | Non-entry into the national phase |
Ref country code: DE |
|
122 | Ep: pct application non-entry in european phase |
Ref document number: 11780534 Country of ref document: EP Kind code of ref document: A1 |