-
公开(公告)号:WO2023015808A1
公开(公告)日:2023-02-16
申请号:PCT/CN2021/139127
申请日:2021-12-17
Applicant: 浙江智慧视频安防创新中心有限公司
IPC: H04N19/109 , H04N19/503
Abstract: 本申请公开了一种数字视网膜视频处理方法、装置、电子设备及存储介质。其中,该方法包括:至少部分基于确定的特征编码参考帧确定帧间预测参考帧;和/或至少部分基于确定的帧间预测参考帧确定特征编码参考帧;在视频压缩编码码流中发送参考帧指示信息;其中,所述参考帧指示信息为所述特征参考帧与所述帧间预测参考帧之间的关联关系。
-
公开(公告)号:WO2023011098A1
公开(公告)日:2023-02-09
申请号:PCT/CN2022/104253
申请日:2022-07-07
Applicant: 腾讯科技(深圳)有限公司
Inventor: 毛峻岭
IPC: H04N21/6375 , H04N19/105 , H04N19/503
Abstract: 本申请公开一种数据传输方法及相关装置,检测到第i个前向参考帧发生丢包,在视频解码过程中,若解码到第i+T+1个前向参考帧,由于第i+T+1个前向参考帧的第一参考帧组合中的视频帧位于第i+T+1个前向参考帧之前,第i个前向参考帧与第i+T+1个前向参考帧之间间隔T个前向参考帧,T是正整数,以便在解码第i+T+1个前向参考帧之前有较多的时间进行第i个前向参考帧的重传,使得在解码到第i+T+1个前向参考帧时完成第i个前向参考帧的重传的可能性提高,降低解码第i+T+1个前向参考帧的可能等待时间。故利用包括重传的第i个前向参考帧的第一参考帧组合对第i+T+1个前向参考帧进行解码。降低接收端视频播放的时延,提高视频播放的流畅性。
-
公开(公告)号:WO2023285469A1
公开(公告)日:2023-01-19
申请号:PCT/EP2022/069479
申请日:2022-07-12
Applicant: TOWARDS CONSEIL
Inventor: BOUILLOUX, Matthieu
IPC: H04N19/63 , H04N19/124 , H04N19/132 , H04N19/137 , H04N19/182 , H04N19/169 , H04N19/503
Abstract: L'invention concerne un procédé d'encodage d'un flux vidéo permettant d'obtenir un flux vidéo à faible latence. Le procédé d'encodage comprend plusieurs étapes, notamment une étape (110) de transformée en ondelettes pour chaque espace de couleur de chaque trame vidéo du flux vidéo, une étape (120) d'approximation de luminance pour permettre la réduction (130) de l'information colorimétrique de chaque information de couleur, une étape de réduction du nombre de données par détection (140) de mouvement par rapport aux frames passées du flux vidéo, puis un codage des pixels à mettre à jour par un tableau de coordonnées, une étape optionnelle de suppression de données par perte d'information à faible impact (valeurs de pixels solitaires, faibles amplitudes, définition), une étape (180) d'identification et notation de bandelettes et de répétition, et une dernière étape (200) de compression sans perte grâce à un algorithme de compression sans perte de type Deflate.
-
公开(公告)号:WO2023283184A1
公开(公告)日:2023-01-12
申请号:PCT/US2022/036111
申请日:2022-07-05
Applicant: GOOGLE LLC
Inventor: TODERICI, George Dan , AGUSTSSON, Eirikur Thor , MENTZER, Fabian Julius , MINNEN, David Charles , BALLE, Johannes , JOHNSTON, Nicholas
IPC: H04N19/117 , H04N19/139 , H04N19/182 , H04N19/537 , G06N3/04 , H04N19/503 , H04N19/513 , H04N19/172 , H04N19/82
Abstract: Methods, systems, and apparatus, including computer programs encoded on a computer storage medium, for compressing video data. In one aspect, a method comprises: receiving a video sequence of frames; generating, using a flow prediction network, an optical flow between two sequential frames, wherein the two sequential frames comprise a first frame and a second frame that is subsequent the first frame; generating from the optical flow, using a first autoencoder neural network: a predicted optical flow between the first frame and the second frame; and warping a reconstruction of the first frame according to the predicted optical flow and subsequently applying a blurring operation to obtain an initial predicted reconstruction of the second frame.
-
公开(公告)号:WO2022252237A1
公开(公告)日:2022-12-08
申请号:PCT/CN2021/098485
申请日:2021-06-04
Applicant: 华为技术有限公司
IPC: H04N19/503 , G06T9/00
Abstract: 本申请提供一种3D地图的编解码方法及装置。本申请3D地图的编码方法,包括:获取3D地图包含的多个3D地图点,多个3D地图点具有跨模态相关性,多个3D地图点具有跨模态相关性是指多个3D地图点中的任意一个3D地图点在多个3D地图点中的顺序与其多个属性相关联,3D地图点的多个属性至少包括3D地图点描述子;根据至少一个参考3D地图点的数据对当前3D地图点的数据进行预测得到当前3D地图点的残差数据,数据至少包括多个属性,多个3D地图点包括当前3D地图点和至少一个参考3D地图点,至少一个参考3D地图点中的每一个参考3D地图点都是在编码当前3D地图点之前已编码的3D地图点;对当前3D地图点的残差数据编码以得到码流。本申请可以提高3D地图的传输效率。
-
6.
公开(公告)号:WO2022225643A1
公开(公告)日:2022-10-27
申请号:PCT/US2022/021870
申请日:2022-03-25
Applicant: TENCENT AMERICA LLC
Inventor: JIANG, Wei , WANG, Wei , LIN, Sheng , XU, Xiaozhong , LIU, Shan
IPC: H04N19/82 , H04N19/109 , H04N19/176 , H04N19/503 , H04N19/80 , G06N3/02 , G06N3/08
Abstract: A method and apparatus of for video enhancement based on neural network based loop filtering using meta learning may include receiving reconstructed video data; receiving one or more quality factors associated with the reconstructed video data; and generating enhanced video data with artefact reduction, based on the one or more quality factors and the reconstructed video data, using a neural network based loop filter, wherein the neural network based loop filter comprises neural network based loop filter parameters that include shared parameters and adaptive parameters.
-
公开(公告)号:WO2022197375A8
公开(公告)日:2022-09-22
申请号:PCT/US2022/014255
申请日:2022-01-28
Applicant: TENCENT AMERICA LLC
IPC: H04N19/00 , H04N19/117 , H04N7/00 , H04N19/105 , H04N19/107 , H04N19/132 , H04N19/14 , H04N19/159 , H04N19/176 , H04N19/186 , H04N19/503 , H04N19/593 , H04N19/70 , H04N19/82
Abstract: A method for in-loop sample offset filtering in a video decoder is disclosed. The method includes obtaining at least one statistical property associated with reconstructed samples of at least a first color component in a current reconstructed data block of a video stream, selecting a target sample offset filter among a plurality of sample offset filters based on the at least one statistical property, the target sample offset filter comprising a nonlinear mapping between sample delta measures and sample offset values, and filtering a current sample in a second color component of the current reconstructed data block using the target sample offset filter and reference samples in a third color component of the current reconstructed data block to generate a filtered reconstructed sample of the current sample.
-
公开(公告)号:WO2022117369A1
公开(公告)日:2022-06-09
申请号:PCT/EP2021/082325
申请日:2021-11-19
Applicant: NOKIA TECHNOLOGIES OY
Inventor: PANUSOPONE, Krit , HONG, Seungwook , WANG, Limin
IPC: H04N19/105 , H04N19/136 , H04N19/503 , H04N19/109 , H04N19/139 , H04N19/52 , H04N19/513 , H04N19/176
Abstract: Methods, apparatuses, and computer program products are provided for content adaptive collocated reference picture selection. Simplified selection criteria are used to determine a level of temporal motion vector prediction (TMVP) support provided by two or more reference pictures from a picture sequence. Selection of a reference picture as collocation reference picture may be based on which of the reference pictures (or a subset of the reference pictures) in the picture sequence, has the most coding blocks for which valid motion vector data is available. Selection may also or alternatively be based on a comparison of a number of inter coded blocks, a number of bi-prediction coded blocks and/or a temporal distance of reference pictures from a current picture of the picture sequence, quantization parameters, a reference list, sequential selection processes, weighted sum values, or the like. Collocation reference picture may be stored in buffer for the current picture.
-
公开(公告)号:WO2021244125A1
公开(公告)日:2021-12-09
申请号:PCT/CN2021/085454
申请日:2021-04-02
Applicant: OPPO广东移动通信有限公司
Inventor: 杨宁
IPC: H04N19/503 , H04N19/513 , H04N19/52
Abstract: 一种帧间预测方法,应用于解码器,所述方法包括:解析码流,获取当前块的权重导出模式;基于所述当前块的权重导出模式,构建新的运动信息候选列表;基于所述新的运动信息候选列表,确定所述当前块的帧间预测值。一种解码器、编码器及计算机存储介质。
-
公开(公告)号:WO2021125751A1
公开(公告)日:2021-06-24
申请号:PCT/KR2020/018369
申请日:2020-12-15
Applicant: 현대자동차주식회사 , 기아자동차주식회사 , 광운대학교 산학협력단
IPC: H04N19/117 , H04N19/82 , H04N19/132 , H04N19/86 , H04N19/119 , H04N19/503 , H04N19/593 , H04N19/176
Abstract: 현재블록의 변환계수들을 역변환하는 방법 및 장치를 개시한다. 본 발명의 일 실시예에 의하면, 현재블록을 제1모드에 따라 예측하는 방법으로서, 분할 모드 신택스 요소에 기초하여, 상기 현재블록을 비직사각형(non-rectangular) 블록들로 분할하는 단계; 상기 비직사각형 블록들 중에서, 인트라 예측되는 인트라 블록 및 인터 예측되는 인터 블록을 결정하는 단계; 및 움직임 정보에 기초하여 상기 인터 블록을 포함하는 제1영역의 예측 샘플들을 유도하고, 인트라 예측 모드에 기초하여 상기 인트라 블록을 포함하는 제2영역의 예측 샘플들을 유도하는 단계를 포함하는 것을 특징으로 하는 방법을 제공한다.
-
-
-
-
-
-
-
-
-