公开(公告)号：WO2023061306A1

公开(公告)日：2023-04-20

申请号：PCT/CN2022/124209

申请日：2022-10-09

申请人： BEIJING BYTEDANCE NETWORK TECHNOLOGY CO., LTD. ,  BYTEDANCE INC.

发明人： SALEHIFAR, Mehdi ,  HE, Yuwen ,  ZHANG, Kai ,  ZHANG, Na ,  ZHANG, Li

IPC分类号： H04N19/137 ,  H04N19/159

摘要： Embodiments of the present disclosure provide a solution for video processing. A method for video processing comprises: reordering, during a conversion between a current video block of a video and a bitstream of the video, a plurality of candidate signs of a motion vector difference (MVD) for a motion candidate of the current video block; and performing the conversion based on the reordered plurality of candidate signs. Compared with the conventional solution, the proposed method can advantageously improve coding efficiency and coding quality.

公开(公告)号：WO2023043824A1

公开(公告)日：2023-03-23

申请号：PCT/US2022/043507

申请日：2022-09-14

申请人： TENCENT AMERICA LLC

发明人： ZHAO, Xin ,  XU, Xiaozhong ,  LIU, Shan

IPC分类号： H04N19/593 ,  H04N19/52 ,  H04N19/105 ,  H04N19/159 ,  H04N19/176

摘要： This disclosure relates generally to video coding and particularly to intra block copy coding mode. For example, a method is disclosed for processing video data which may include receiving a video bitstream comprising at least one block of a video frame, the at least one block comprising a current block; determining a first search area in the video frame; determining a second search area, wherein the second search area is a second candidate area for locating the IntraBC prediction block; identifying the IntraBC prediction block referenced by a block vector; determining a prediction based on a spatial relationship between the IntraBC prediction block, the first search area, and the second search area; and decoding the at least the portion of the current block based on the determined prediction.

公开(公告)号：WO2023024115A1

公开(公告)日：2023-03-02

申请号：PCT/CN2021/115154

申请日：2021-08-27

申请人： OPPO广东移动通信有限公司

发明人： 李帅 ,  席欣 ,  高艳博 ,  元辉 ,  李明

IPC分类号： H04N19/159 ,  H04N19/172

摘要： 本申请实施例提供了一种编码方法、解码方法、编码器、解码器和解码系统。该编码方法包括：获取第一图像中的编码块的第一划分信息；根据第一图像和第一划分信息，对机器学习模型进行更新，其中，该机器学习模型是根据编码块样本进行预先训练得到的，该编码块样本包括编码块的像素信息和划分信息；将第二图像输入更新后的机器学习模型，以得到第二图像的编码块的第二划分信息，并根据该第二划分信息，对第二图像进行编码。本申请实施例通过对已经训练好的深度学习模型进行更新，能够使得深度学习模型对于第一图像所在的图像序列达到更好的拟合度，从而有助于寻找到当前模型对当前测试序列的全局最优解，有利于进一步提高机器学习模型的准确率。

公开(公告)号：WO2023013667A1

公开(公告)日：2023-02-09

申请号：PCT/JP2022/029748

申请日：2022-08-03

申请人： ＫＤＤＩ株式会社

发明人： 木谷　佳隆

IPC分类号： H04N19/105 ,  H04N19/159 ,  H04N19/176 ,  H04N19/50

摘要： 本発明に係る画像復号装置２００は、ＧＰＭに対する動き情報を導出して動き補償画素を生成するインター予測部２４１と、ＧＰＭに対するイントラ予測モードを導出してイントラ予測画素を生成するイントラ予測部２４２と、ＧＰＭが適用された復号対象ブロックの動き情報又はイントラ予測モードと、インター予測及びイントラ予測のいずれかが適用されたかを判定可能な予測種別とを含む予測情報を保存又は出力する予測情報バッファ２４４とを備える。

公开(公告)号：WO2023275434A1

公开(公告)日：2023-01-05

申请号：PCT/FI2022/050393

申请日：2022-06-07

申请人： NOKIA TECHNOLOGIES OY

发明人： GHAZNAVI YOUVALARI, Ramin ,  ASTOLA, Pekka

IPC分类号： H04N19/186 ,  H04N19/192 ,  H04N19/105 ,  H04N19/176 ,  H04N19/593 ,  H04N19/159

摘要： A method comprising: obtaining an image block unit comprising samples in color channels of one or two chrominance channels and one luminance channel (800); reconstructing samples of said luminance channels of the image block unit (802); predicting samples of at least one color channel of image block unit using a cross-component prediction model based on reconstructed samples on a reference channel, wherein parameters of the cross-component prediction model are determined at least partly based on a plurality of reconstructed chrominance samples and their corresponding reconstructed luminance samples in neighboring image block units (804); predicting a plurality of color channel samples in said neighboring image block units co-locating at least part of the reconstructed chrominance and luminance samples using said cross-component prediction model (806); determining a difference between values of the predicted plurality of color channel samples in said neighboring image block and corresponding reconstructed color channel samples (808); and adjusting the parameters of the cross-component prediction model based on said difference (810).

公开(公告)号：WO2022257970A1

公开(公告)日：2022-12-15

申请号：PCT/CN2022/097634

申请日：2022-06-08

申请人： 维沃移动通信有限公司

发明人： 张伟 ,  聂佳明 ,  吕卓逸 ,  杨付正

IPC分类号： G06T17/20 ,  H04N19/159

摘要： 本申请公开了一种点云的几何信息编码处理方法、解码处理方法及相关设备，本申请实施例的点云的几何信息编码处理方法包括：对第一几何信息和第二几何信息进行同步树划分，获得第一几何信息对应的第一占位码和第二几何信息对应的第二占位码；第一几何信息基于当前待编码的第N帧点云的几何信息进行预处理得到，N为大于1的整数，第二几何信息为第N帧点云对应的参考帧点云的重建几何信息；根据第一节点和第一节点的邻居节点的占据情况，确定目标上下文；第一节点为第二几何信息对应的树结构中与第二节点对应的节点，第二节点为第一几何信息对应的树结构中当前待编码节点中待编码子节点；利用目标上下文对第二节点的占位码进行熵编码。

公开(公告)号：WO2022211715A1

公开(公告)日：2022-10-06

申请号：PCT/SE2022/050315

申请日：2022-03-30

申请人： TELEFONAKTIEBOLAGET LM ERICSSON (PUBL)

发明人： ANDERSSON, Kenneth ,  ENHORN, Jack

IPC分类号： H04N19/14 ,  H04N19/124 ,  H04N19/176 ,  H04N19/159 ,  H04N19/51

摘要： A method for making a smooth prediction block of samples in a picture in an image or video encoder or decoder is provided. The method includes determining parameters r of a polynomial model by r = (BTB)-1*(BT*x), where B is a matrix of bases and x is a source of samples in vector form. The method includes predicting a block x' based on the parameters r and the matrix of bases B.

公开(公告)号：WO2022196133A1

公开(公告)日：2022-09-22

申请号：PCT/JP2022/003498

申请日：2022-01-31

申请人： キヤノン株式会社

发明人： 堀下　祐輔

IPC分类号： H04N19/105 ,  H04N19/107 ,  H04N19/146 ,  H04N19/147 ,  H04N19/159 ,  H04N19/176

摘要： 入力画像のブロック毎に符号化処理を行う符号化装置であって、インター予測、イントラ予測、マルチ参照予測のうちいずれかの予測モードを用いてブロックごとに符号化処理を実行する符号化手段と、符号化処理の結果を符号化データとして出力する出力手段と、ブロックをインター予測モードで符号化した場合の第１の符号化コストおよびイントラ予測モードで符号化した場合の第２の符号化コストを所定の関数を用いて算出する算出手段と、を備え、符号化手段は、第１の符号化コストと第２の符号化コストの差分が閾値以下の場合、マルチ参照予測モードを用いて符号化処理を実行し、差分が閾値より大きい場合、第１の符号化コストと第２の符号化コストのうち小さい方の予測モードを用いて符号化処理を実行する。

公开(公告)号：WO2022149156A1

公开(公告)日：2022-07-14

申请号：PCT/IN2021/051172

申请日：2021-12-15

申请人： SLING MEDIA PVT LTD.

发明人： VENSIYANI, Abhilash Magan ,  RAIKAR, Yatish Jayant Naik ,  TRIPATHI, Varunkumar B. ,  DEVARAJ, Vivek

IPC分类号： H04N21/2343 ,  H04N19/159 ,  H04N19/172 ,  H04N19/44 ,  H04N19/70

摘要： Systems and processes are provided to generate a thumbnail image including a tuner configured for receiving and demodulating a video signal including a video packet having a packet header and a video frame, a processor configured for generating a modified packet header by modifying the packet header to set a number of reference frames to zero and to designate the video frame as unused for reference, decoding the video frame in response to the modified packet header to extract the video frame, and generating the thumbnail image in response to the video frame, a memory configured for storing the thumbnail image, and a video output configured for coupling the thumbnail image to a display in response to a user video search request.

公开(公告)号：WO2022125160A1

公开(公告)日：2022-06-16

申请号：PCT/US2021/048624

申请日：2021-09-01

申请人： TENCENT AMERICA LLC [US]/[US]

发明人： XU, Xiaozhong ,  LIU, Shan

IPC分类号： H04N19/159 ,  H04N19/176 ,  H04N19/186 ,  H04N19/44 ,  H04N19/46

摘要： Aspects of the disclosure provide a method and an apparatus including processing circuitry for video decoding. The processing circuitry can decode coded information for a current block in a current picture from a coded video bitstream. The coded information can indicate a string copy mode for the current block. The current block includes a current string. The processing circuitry can determine whether a flipping operation is performed to predict the current string. Based on the flipping operation being determined as performed to predict the current string, the processing circuitry can determine an original reference string based on a string vector of the current string. The processing circuitry can generate a flipped reference string by performing the flipping operation on the original reference string, and reconstruct the current string based on the flipped reference string.