Apparatuses, systems, and techniques to interpolate one or more intermediate images from two or more images is disclosed. In at least one embodiment, a processor includes one or more circuits to interpolate one or more intermediate images from two or more images based, at least in part, on one or more inconsistent flow vectors corresponding to the two or more images.

A method for processing a video includes performing a conversion between a current block of visual media data and a corresponding coded representation of the visual media data, wherein the conversion of the current block includes determining whether a use of one or both of a bi-directional optical flow (BIO) technique or a decoder-side motion vector refinement (DMVR) technique to the current block is enabled or disabled, and wherein the determining the use of the BIO technique or the DMVR technique is based on a cost criterion associated with the current block.

This disclosure relates generally to video coding/decoding and particularly to methods and systems for refining motion vector candidates. An example decoding method includes receiving a coded video bitstream; extracting, from the coded video bitstream, a first syntax element indicating whether a compound reference mode is applied for a current video block; extracting, from the coded video bitstream, a second syntax element indicating whether a motion vector difference (MVD) is joined signaled for the current video block; in response to the first syntax element indicating that the compound reference mode is applied for the current video block and the second syntax element indicating that the MVD is joined signaled for the current video block, refining, by the device, at least one MV predictor (MVP) in a MV candidate list based on whether the at least one MVP is symmetric; and decoding, by the device based on the MV candidate list, the current video block.

A method and a device for encoding or decoding based on an inter-frame prediction. The method includes steps of: determining a temporal motion vector prediction value of a to-be-processed coding unit, where the temporal motion vector prediction value is a temporal motion vector prediction value of a sub-block, a temporal motion vector of which is obtainable through prediction, in sub-blocks adjacent to the to-be-processed coding unit and/or sub-blocks in the to-be-processed coding unit; determining a motion vector residual prediction value of the to-be-processed coding unit according to the temporal motion vector prediction value; determining a motion vector of a sub-block in the to-be-processed coding unit according to the temporal motion vector prediction value and the motion vector residual prediction value and performing a motion compensation according to the motion vector of the sub-block in the to-be-processed coding unit to determine a prediction block of the to-be-processed coding unit.

A method of partitioning a video coding block for JVET, comprising representing a JVET coding tree unit as a root node in a quadtree plus binary tree (QTBT) structure that can have a quadtree branching from the root node and binary trees branching from each of the quadtree's leaf nodes using asymmetric binary partitioning to split a coding unit represented by a quadtree leaf node into two child nodes of unequal size, representing the two child nodes as leaf nodes in a binary tree branching from the quadtree leaf node and coding the child nodes represented by leaf nodes of the binary tree with JVET, wherein weighted angular coding unit prediction is employed and wherein post-coding unit prediction filtering can be avoided or unbiased to increase coding times.

An encoding method, playing method and apparatus for image stabilization of panoramic video, and a method for evaluating image stabilization algorithm are provided. The image stabilization method for the panoramic video is applicable to an electronic apparatus including a processor. In the method, a plurality of image frames of a panoramic video is captured, and each image frame is transformed into a plurality of projection frames on a plurality of faces of a cubemap. Then, variations of triaxial displacements and attitude angles between the projection frames transformed onto each of the faces and adjacent in time are calculated. The variations of triaxial displacements and attitude angles are smoothed and recorded as movement information. While playing the panoramic video, the panoramic video is corrected by the movement information and played. Thus, it is possible to reduce the amount of calculation required for the stabilization calculations on the captured video.

The present disclosure relates to an image processing device and an image processing method capable of suppressing deterioration of image quality and deterioration of encoding efficiency.

An image processing device includes: an inter-prediction unit performs motion compensation processing to which optical flow processing is applied on a color difference component of a current prediction block that is subject to an encoding process as color difference optical flow processing to generate a prediction pixel in the current prediction block; and an encoding unit that encodes a current pixel in the current prediction block using the prediction pixel. The present technology can be applied to, for example, an image processing system that performs encoding and decoding according to the VVC method.

An image decoding device includes: a motion vector decoding unit that decodes a motion vector from encoded data; a refinement unit that performs refinement processing to correct the decoded motion vector; and a predictive signal generation unit that generates a predictive signal on the basis of the corrected motion vector outputted from the refinement unit. The predictive signal generation unit determines whether or not to apply BDOF processing for each block, on the basis of information calculated in the course of the refinement processing.

A prediction image correcting device comprises a predictor (108) configured to predict a target image block obtained by dividing a present image frame by using a plurality of reference images to generate a prediction image corresponding to the target image block a prediction accuracy evaluator (109) configured to evaluate prediction accuracy of the prediction image based on a degree of similarity between the plurality of reference images used for generating the prediction image and a corrector (110) configured to perform correction processing on the prediction image by using a decoded neighboring block adjacent to the target image block, wherein the corrector is configured to control the correction processing based at least on an evaluation result of the prediction accuracy evaluator.

A method for estimating a motion vector of a pixel block, a video processing apparatus, an electronic device and a storage medium. The method for estimating the motion vector of the pixel block includes: obtaining a plurality of candidate motion vectors of a current pixel block in a current video frame, the plurality of candidate motion vectors comprising at least a down-sampled candidate motion vector; calculating a confidence degree for a difference value between each of other candidate motion vectors except for the down-sampled candidate motion vector in the plurality of candidate motion vectors and the down-sampled candidate motion vector; and determining one candidate motion vector in the other candidate motion vectors as the motion vector of the current pixel block, the confidence degree for the difference value between the one candidate motion vector and the down-sampled candidate motion vector is highest.