公开(公告)号：WO2019118026A1

公开(公告)日：2019-06-20

申请号：PCT/US2018/051036

申请日：2018-09-14

Applicant: GOOGLE LLC

Inventor： HAN, Jingning ,  HE, Dake

IPC: H04N19/176 ,  H04N19/129

Abstract: A scan order for encoding or decoding coefficients of a transform block is selected on a transform block-level. A set of candidate scan orders is processed by identifying end of block positions within the transform block for candidate scan orders. Cost values are determined for each of the candidate scan orders to reflect a number of the coefficients of the transform block that are located before the respective end of block positions. In particular, a cost value for a candidate scan order reflects the number of zero-value coefficients located before the end of block position for that candidate scan order. One of the candidate scan orders is then selected based on those cost values. The selected scan order is used to scan the coefficients in the transform block, such as for encoding those coefficients to a bitstream or for decoding those coefficients to an output video stream.

公开(公告)号：WO2019013842A1

公开(公告)日：2019-01-17

申请号：PCT/US2018/022794

申请日：2018-03-16

Applicant: GOOGLE LLC

Inventor： CHIANG, Ching-Han ,  HAN, Jingning ,  XU, Yaowu

IPC: H04N19/96 ,  H03M7/40

Abstract: Video syntax elements are coded using a context tree. Context information used for coding previously-coded syntax elements is identified. A context tree is produced by separating the previously-coded syntax elements into data groups based on the context information. The context tree includes nodes representing the data groups. Separating the previously-coded syntax elements can include applying separation criteria against values of the context information to produce at least some of the nodes. Context information is then identified for another set of syntax elements to be coded. One of the nodes of the context tree is identified based on values of the context information associated with one of the other set of syntax elements. That syntax element is then coded according to a probability model associated with the identified node. The context tree can be used to encode or decode syntax elements.

公开(公告)号：WO2018160231A1

公开(公告)日：2018-09-07

申请号：PCT/US2017/059272

申请日：2017-10-31

Applicant: GOOGLE LLC

Inventor： XU, Yaowu ,  HAN, Jingning ,  CHIANG, Ching-Han

IPC: H04N19/463 ,  H04N19/12 ,  H04N19/137 ,  H04N19/176

Abstract: Transform kernel candidates including a vertical transform type associated with a vertical motion and a horizontal transform type associated with a horizontal motion can be encoded or decoded. During an encoding operation, a residual block of a current block is transformed according to a selected transform kernel candidate to produce a transform block. A probability model for encoding the selected transform kernel candidate is then identified based on neighbor transform blocks of the transform block. The selected transform kernel candidate is then encoded according to the probability model. During a decoding operation, the encoded transform kernel candidate is decoded using the probability model. The encoded transform block is then decoded by inverse transforming dequantized transform coefficients thereof according to the decoded transform kernel candidate.

公开(公告)号：WO2020176142A1

公开(公告)日：2020-09-03

申请号：PCT/US2019/059026

申请日：2019-10-31

Applicant: GOOGLE LLC

Inventor： CHEN, Cheng ,  HAN, Jingning

IPC: G06T1/60

Abstract: A memory-efficient filtering approach codes images and video. A buffer having a fixed size based on a size of processing units for filtering a video frame is allocated. For each of the processing units, pre-filtered pixel values are copied from a respective region of the video frame to the buffer based on a writing point for the video frame and an offset applied to the writing point, filtering is performed against the pre-filtered pixel values from the buffer to produce filtered pixel values, and the filtered pixel values are written to the video frame based on the writing point and the offset. The filtering may be performed using a loop restoration tool, such as where the pre-filtered pixel values are output from a constrained directional enhancement filter (CDEF) tool. Alternatively, the filtering may be performed using the CDEF tool or another coding tool.

公开(公告)号：WO2019036075A1

公开(公告)日：2019-02-21

申请号：PCT/US2018/030347

申请日：2018-05-01

Applicant: GOOGLE LLC

Inventor： HAN, Jingning ,  XU, Yaowu

IPC: H04N19/577 ,  H04N19/567

Abstract: A prediction scheme is selected for encoding or decoding a video block. A first compound motion block can be determined by weighting distances from a first reference frame to the video frame and from a second reference frame to the video frame using one or more quantized weighting coefficients. A second compound motion block can be determined based on an average of pixel values a video block of the first reference frame and pixel values from a video block of the second reference frame. One of the first compound motion block or the second compound motion block is selected and used to generate a prediction block. Alternatively, data encoded to a bitstream including the video frame can be used to determine which compound motion block to use to generate the prediction block. The current block of the video frame is then encoded or decoded using the prediction block.

公开(公告)号：WO2018118134A1

公开(公告)日：2018-06-28

申请号：PCT/US2017/044649

申请日：2017-07-31

Applicant: GOOGLE LLC

Inventor： MUKHERJEE, Debargha ,  XU, Yaowu ,  BANKOSKI, James ,  WILKINS, Paul ,  HAN, Jingning ,  LIU, Yuxin ,  LIN, Wei-Ting

IPC: H04N19/134 ,  H04N19/117 ,  H04N19/82

Abstract: Encoding or decoding blocks of video frames using multiple reference frames with adaptive temporal filtering can include generating one or more candidate reference frames by applying temporal filtering to one or more frames of a video sequence according to relationships between respective ones of the one or more frames and a current frame of the video sequence. A reference frame to use for predicting the current frame can be selected from the one or more candidate reference frames, and a prediction block can be generated using the selected reference frame. During an encoding operation, the prediction block can be used to encode a block of a current frame of the video sequence. During a decoding operation, the prediction block can be used to decode a block of a current frame of the video sequence.

公开(公告)号：WO2018048516A1

公开(公告)日：2018-03-15

申请号：PCT/US2017/044639

申请日：2017-07-31

Applicant: GOOGLE LLC

Inventor： HAN, Jingning ,  CHIANG, Ching-Han

IPC: H04N19/176 ,  H04N19/134 ,  H04N19/129 ,  H04N19/60 ,  H04N19/136 ,  H04N19/18 ,  H04N19/192

Abstract: Video coding using a context adaptive scan order for entropy coding may include decoding a current block by identifying a context adaptive scan order for entropy decoding a transform block, which may include identifying non-zero-coefficient probabilities for the transform block such that each location in the transform block corresponds to a respective non-zero-coefficient probability from the non-zero-coefficient probabilities, assigning a respective context adaptive scan order position to each location in the transform block in descending magnitude order of the respective corresponding non-zero-coefficient probabilities such that the context adaptive scan order position for each location exceeds the context adaptive scan order position assigned to entropy coding context locations for the respective location, entropy decoding transform coefficients from the encoded video stream based on the context adaptive scan order, and reconstructing the decoded block based on the transform block.

Abstract translation: 使用用于熵译码的上下文自适应扫描次序的视频译码可包含通过识别用于对变换块进行熵译码的上下文自适应扫描次序来解码当前块，所述上下文自适应扫描次序可包括将用于熵解码的非零系数概率 所述变换块使得所述变换块中的每个位置对应于来自所述非零系数概率的相应的非零系数概率;以相应的上下文自适应扫描顺序位置向所述变换块中的每个位置分配按照 各个对应的非零系数概率使得每个位置的上下文自适应扫描次序位置超过分配给各个位置的熵译码上下文位置的上下文自适应扫描次序位置，基于所述编码视频流对来自编码视频流的变换系数进行熵解码 上下文自适应扫描顺序，并且基于tra重建解码块 nsform block。

公开(公告)号：WO2021092531A1

公开(公告)日：2021-05-14

申请号：PCT/US2020/059594

申请日：2020-11-09

Applicant: GOOGLE LLC

Inventor： HAN, Jingning ,  SUN, Yue ,  XU, Yaowu

IPC: H04N19/13 ,  H04N19/91 ,  H03M7/40

Abstract: Entropy coding a sequence of symbols is described. A first probability model for entropy coding is selected. At least one symbol of the sequence is coded using a probability determined using the first probability model. The probability according to the first probability model is updated with an estimation of a second probability model to entropy code a subsequent symbol. The combination may be a fixed or adaptive combination.

公开(公告)号：WO2019099084A1

公开(公告)日：2019-05-23

申请号：PCT/US2018/047223

申请日：2018-08-21

Applicant: GOOGLE LLC

Inventor： XU, Yaowu ,  HAN, Jingning ,  CHIANG, Ching-Han

IPC: H04N19/176 ,  H04N19/70 ,  H04N19/147 ,  H04N19/567 ,  H04N19/577 ,  H04N19/51

CPC classification number: H04N19/573 ,  H04N19/147 ,  H04N19/176 ,  H04N19/182 ,  H04N19/44 ,  H04N19/51 ,  H04N19/567 ,  H04N19/577 ,  H04N19/70

Abstract: A motion-field based reference frame is rendered that can be used to perform a prediction process for a first frame of a video sequence to be predicted. A first reference frame from the video sequence for forward inter prediction of the first frame is determined, as is a second reference frame from the video sequence for backward inter prediction of the first frame. A respective motion field determined for blocks of the first frame is used to determine parameters forming an affine homographic model. A current block of a reconstructed reference frame is rendered at a co-located position within a motion field-based reference frame by applying the affine homographic model to the current block. An affine homographic model may be determined for each reconstructed reference frame block to render the motion-field based reference frame.

公开(公告)号：WO2019055070A1

公开(公告)日：2019-03-21

申请号：PCT/US2018/031565

申请日：2018-05-08

Applicant: GOOGLE LLC

Inventor： HAN, Jingning ,  XU, Yaowu

IPC: H04N19/82 ,  H04N19/86 ,  H04N19/176 ,  H04N19/14 ,  H04N19/122

Abstract: Video coding using dual deblocking filter thresholds include generating a reconstructed frame by decoding an encoded bitstream and outputting the reconstructed frame. Decoding may include generating a decoded block by decoding a portion of the encoded bitstream, identifying a first deblocking threshold index from the encoded bitstream, identifying a second deblocking threshold index from the encoded bitstream, generating a reconstructed block based on the decoded block, and including the reconstructed block in the reconstructed frame. Generating the reconstructed block may include deblocking based on the first deblocking threshold index and the second deblocking threshold index.