Canon Kabushiki Kaisha

Interdigital Vc Holdings, Inc.

Interdigital Ce Patent Holdings, Sas

Interdigital Madison Patent Holdings, Sas

Thomson Licensing

Canon Research Centre France

Fabrice Le Leannec

Patrice Onno

Xavier Henocq

Franck Galpin

Tangi Poirier

Philippe Bordes

Sebastien Lasserre

Franck Denoual

Fabien Racape

Karam Naser

Gagan Bihari Rath

Naël Ouedraogo

Yannick Olivier

David Touze

Ya Chen

Thierry Dumas

Edouard Francois

Fabrice Urban

Philippe De Lagrange

Ioana Donescu

Antoine Robert

Sébastien Lasserre

Sunil Jaiswal

Pierre Andrivon

Hervé Ruellan

Eric Nassor

Catherine Serre

Gaëlle Martin-Cocher

Raphaël Grosbois

In one implementation, the CCB counting methods are unified between the transform residual coding process and Transform Skip (TS) residual coding process. In one example, in TS residual coding, the CCB counting excludes the coeff_sign_flag so that syntax used for CCB count is unified for the two residual coding processes. In addition, a separate maximum number of context coded bins can be specified and used for coeff_sign_flag only. In another example, in TS residual coding, the maximum number of CCB count is reduced from TB_size*2 for a TB to TB_size*1.75, or more generally both the maximum CCB counts of the transform residual coding and TS residual coding are set to an identical value, so that the maximum CCB count is unified for the two residual coding processes.

Unification of context-coded bins (CCB) count method

Decoder side intra mode derivation (DIMD) capability is enhanced by using reference pixels that extend several rows above and several columns left of a current video block, and also includes pixels above and left of the current video block, as well as columns above and right and rows below and left of the current video block. The reference pixels are formed from surrounding reconstructed/previously encoded samples surrounding the current video block. The derivation of an intra prediction mode is determined from gradients determined from each of the reference pixels in a defined surrounding area. In one embodiment, the gradients are determined using horizontal and vertical filters. In a sub-embodiment, the filters do not extend beyond the defined surrounding area. In another embodiment, reassignment of an index indicative of the target intra prediction mode is performed.

Improving the angle discretization in decoder side intra mode derivation

A reconstruction method comprising: obtaining () a region predictor for a current region of a picture based on a spatial prediction; obtaining () an information representative of a partitioning of the current region into sub-regions; reconstructing () at least one residual sub-region based on an intra prediction residual and an intra prediction mode and obtaining a reconstructed residual region from the at least one reconstructed residual sub-region, wherein all reference samples external to the current region used for reconstructing the at least one sub-region are set to zero before reconstructing the at least one residual sub-region; and, reconstructing () the current region based on the reconstructed residual region and on the region predictor by adding a final predictor obtained from the region predictor and the reconstructed residual region.

Spatial illumination compensation on large areas

A method for reconstructing a block of a picture comprising: applying an inverse transform to a block of transform coefficients using a first matrix of integer coefficients representative of the inverse transform, wherein the first matrix of integer coefficients is the inverse of a second matrix of integer coefficients representative of a transform inverted to obtain the inverse transform, the second matrix of integer coefficients being derived from a third matrix of integer coefficients, the absolute values of the integer coefficients of the third matrix taking their values in a set of four different values, the derivation comprising a modification of one value of the set to ensure that a sum of two values of the set equals a third value of the set.

High precision 4×4 DST7 and DCT8 transform matrices

At least a method and an apparatus are presented for efficiently encoding or decoding video. For example, an intra prediction of an image block using at least one neural network from a context comprising pixels surrounding the image block is determined and an information relative to a transform method to apply for decoding the image block is also determined. The transform method is adapted to the neural network intra prediction mode of the block to encode or decode. The information relative to the transform method is inferred from the at least one neural network used in intra prediction of the image block at the encoding and either signaled or also inferred at the decoding.

Adapting the transform process to neural network-based intra prediction mode

A method for controlling a display of a frame representing a state of a game in a network-based gaining application comprising: sending (A) an information representative of a user action to a server; receiving (B) at least one frame representing a predicted state of the game predicted from the user action; obtaining an information allowing determining if said at least one frame corresponds to the user action; and; if the at least one frame corresponds to said user action, determining when displaying one of the at least one frame in function of a time at which this user action was performed and of said information.

Display control in cloud gaming applications

Video encoding and decoding using non-rectangular partitioning. An example video decoding method includes: obtaining information identifying a non-rectangular first partitioning of a current block into a plurality of prediction units; determining a second partitioning of the current block into a plurality of transform units, wherein each transform unit is completely spatially contained within a respective one of the prediction units; and decoding the current block according to the non-rectangular first partitioning and the second partitioning. In some embodiments, the scan order of the transform units is determined such that all transform units contained in a same one of the prediction units are scanned before starting to scan transform units that are contained in a different one of the prediction units.

Method and device for intra-predictive encoding/decoding a coding unit comprising picture data, said intra-predictive encoding depending on a prediction tree and a transform tree

A method for video decoding ensuring a consistent interaction of some signaling tools with some encoding tools. The method is particularly advantageous when a container intended to provide parameters required for applying an encoding tool is missing while the encoding tool is activated for a given block. Examples include obtaining a first information indicating if a container of a first type providing at least one coding parameter is absent from the bitstream; checking a value of a first syntax element indicating if a coding tool using said at least one coding parameter is activated for a current block of samples of an image of the video sequence; and, adapting a decoding of the current block in case of absence of the container when the second syntax element indicates an activation of the coding tool.

Method and device for consistent interaction of signaling tools with encoding tools in video decoding

In some coding modes, such as the merge mode, a list of motion candidates is generated using spatial or temporal neighboring blocks or constructed ones, and one of the motion candidates is inherited by the current block. When constructing the list of motion candidates, motion information of a candidate to be added is checked to limit the redundancy in the constructed list. In one implementation, the switchable interpolation filter, the IFindex becomes part of the motion information to be used in comparison since it can modify the resulting prediction. Furthermore, the GBiIdx indicating the weights used in bi-prediction can also be part of the motion information to be compared and could also be taken into account in the redundancy check. In general, all or part of motion information that causes different prediction can be used in redundancy check.

Extended motion information comparison

Decoding or encoding picture information can involve determining a value associated with a prediction of a pixel of a block of picture information, wherein, the prediction is based on an intra-prediction along a non-diagonal direction, determining, from among a plurality of decoded pixels neighboring the block on the top and the left, at least four reference samples for the block based on the non-diagonal direction, and wherein the determining of the value comprises an interpolation based on a quadratic model and the four reference samples; and decoding or encoding at least a portion of the picture information based on the value.

Intra prediction for video encoding and decoding

In one implementation, KLT transform matrices are derived during the encoding or decoding process. In particular, the KLT transform matrices can be derived for either the primary transforming stage or the secondary transforming stage, or both. Because the KLT transform matrices can be derived at both the encoder and decoder sides, the KLT transform matrices do not need to be signaled in the bitstream. To derive the KLT transform matrix for a current block to be encoded or decoded, a template for the current block is defined, and is used to search in reconstructed regions multiple blocks with templates similar to the current template. Those multiple blocks are used to train the KLT transform matrix.

Karhunen loeve transform for video coding

A block of video data is split using one or more of several possible partition operations by using the partitioning choices obtained through use of a texture-based image partitioning. In at least one embodiment, the block is split in one or more splitting operations using a convolutional neural network. In another embodiment, inputs to the convolutional neural network come from pixels along the block's causal borders. In another embodiment, boundary information, such as the location of partitions in spatially neighboring blocks, is used by the texture analysis. Methods, apparatus, and signal embodiments are provided for encoding.

Texture-based partitioning decisions for video compression

Video processing such as encoding and/or decoding a picture can involve deriving an inter-prediction parameter based on first and second merge candidates used to generate a pairwise merge candidate, wherein the inter-prediction parameter comprises at least one of an index for weighted bi-prediction or an interpolation filter index; and encoding at least a portion of the picture information based on the inter-prediction parameter.

Inter-prediction parameter derviation for video encoding and decoding

In general, encoding or decoding a picture part can involve a first CABAC probability model associated with a first flag indicating use of an affine mode and a second CABAC probability model associated with a second flag indicating use of either the affine mode or a second mode different from the affine mode, where the first and second CABAC probability models are different and are determined independently.

Affine mode signaling in video encoding and decoding

At least a method and an apparatus are presented for encoding or decoding video and can involve, for example, obtaining a group of coding units including two or more of a plurality of coding units divided from a current block wherein the two or more of the plurality of coding units share a coding parameter, and the group of coding units overlaps at least two different pipeline units associated with a pipelined decoding operation, and encoding or decoding the current block based on the group of coding units, and the shared coding parameter.

Growing community of inventors

Mouaze, France

Fabrice Le Leannec