A video coder may be configured to determine a partitioning for a block of video data using geometric partitioning mode; construct two uni-prediction motion vector candidate lists for the block of video data, and code the block of video data using uni-prediction based on at least one of the two uni-prediction motion vector candidate lists to generate a decoded block of video data.

An image encoding and decoding method, includes: determining location information of a target reconstructed image block of a current to-be-encoded image block, where the target reconstructed image block is a reconstructed image block used to determine motion information of the current to-be-encoded image block; determining a first transform core pair based on the location information of the target reconstructed image block; and transforming a residual signal of the current to-be-encoded image block based on the first transform core pair, to obtain a transform coefficient.

Poses of a person depicted within video frame may be determined. The poses of the person may be used to generate intermediate video frames between the video frames.

Poses of a person depicted within video frame may be determined. The poses of the person may be used to generate intermediate video frames between the video frames.

Methods, apparatus, systems for video processing, including video encoding or video decoding are described. One example method includes performing a conversion between a video including a video picture that includes one or more subpictures and a bitstream of the video. The bitstream conforms to a format rule specifying that a first syntax element indicating whether an operation is performed across a boundary of a subpicture in the coded layer video sequence is selectively included responsive to a number of subpictures in the video picture.

A method of video decoding performed in a video decoder includes receiving a coded video bitstream including a current block that is divided into a plurality of sub-blocks. The method includes performing sub-block based affine motion compensation on the current block to generate a sub-block prediction for each pixel in each sub-block of the current block. The method further includes determining one or more spatial gradients for each sub-block prediction. The method includes performing, for each sub-block prediction, prediction refinement with an optical flow process using the respective determined one or more spatial gradients and at least one constraint included in the coded video bitstream. The method further includes adding, for each sub-block prediction, an output of the respective prediction refinement to the respective sub-block prediction to generate a final prediction for each pixel in each sub-block of the current block.

There is provided a method for estimating weighted prediction parameters intended to be used for predicting an image block.

Methods and devices for decoding a video stream are provided, a method includes storing previously decoded pictures of the video stream in a decoded picture buffer, including a plurality of first pictures of a same temporal sub-layer, the plurality of first pictures including at least one sub-layer reference picture for predicting a current picture of the video stream; identifying a network abstract layer (NAL) unit type of a picture of the plurality of first pictures; removing, based on the NAL unit type of the picture identified, the picture from the decoded picture buffer; and decoding the current picture using the decoded picture buffer. The decoding includes predicting the current picture using one or more of the at least one sub-layer reference picture that is stored within the decoded picture buffer, after removing the picture from the decoded picture buffer.

At least a method and an apparatus are presented for efficiently encoding or decoding video. For example, a plurality of frames is provided to a motion estimator to produce an output comprising estimated motion information. The estimated motion information is provided to an auto-encoder or an auto-decoder to produce an output comprising reconstructed motion field. The reconstructed motion field and one or more decoded frames of the plurality of frames are provided to a deep neural network to produce an output comprising refined bi-directional motion field. The video is encoded or decoded based on the refined bi-directional motion field.

At least a method and an apparatus are presented for efficiently encoding or decoding video. For example, a plurality of frames is provided to a motion estimator to produce an output comprising estimated motion information. The estimated motion information is provided to an auto-encoder or an auto-decoder to produce an output comprising reconstructed motion field. The reconstructed motion field and one or more decoded frames of the plurality of frames are provided to a deep neural network to produce an output comprising refined bi-directional motion field. The video is encoded or decoded based on the refined bi-directional motion field.