Tencent America LLC

Hfi Innovation Inc.

Mediatek Corporation

Huawei Technologies Co., Limited

Future Wei Technologies, Inc.

Mediatek Singapore Pte. Ltd.

Tencent Technology (Shenzhen) Company Limited

At&t Intellectual Property I, LP

Other

At+t Corp.

Shan Liu

Xin Zhao

Xiaozhong Xu

Xiang Yang Li

Liang Zhao

Stephan Wenger

Byeongdoo Choi

Guichun Li

Xiang Li

Xiang Zhang

Jing Ye

Madhu Peringassery Krishnan

Wen Gao

Wei Jiang

Lien-Fei Chen

Sehoon Yea

Yixin Du

Han Gao

Jun Tian

Tzu-Der Chuang

Meng Xu

Wei W Wang

Chao Huang

Ling Li

Arash Vosoughi

Cheung Auyeung

Wei Wang

Ding Ding

Wang-Lin Lai

Shaw-Min Lei

Shuai Zhao

Ching-Yeh Chen

Yu-Wen Huang

Yu-Chen Sun

Jungsun Kim

Weiwei Feng

Sheng Lin

Wang Lin Lai

Lulin Chen

Haitao Yang

Roman Chernyak

Zhenzhong Chen

Iraj Sodagar

Thuong Nguyen Canh

Yiming Li

Chih-Wei Hsu

Yin Zhao

Mei Guo

Zeqiang Li

Bing Jian

David Crawford Gibbon

Andrea Basso

Yi-Wen Chen

Xin Ru Wang

Hans Peter Graf

Eric Cosatto

Biao Wang

Ximin Zhang

Han Huang

Xiang Ma

Shan Gao

Ding-Yun Chen

Hong Zhang

Jiantong Zhou

Jiali Fu

Yu Wang

Anish Tamse

Huanbang Chen

Bin Zhu

Yingbin Wang

Semih Esenlik

Anand Meher Kotra

Zhijie Zhao

Jicheng An

Sergey Yurievich Ikonin

Shih-Ta Hsiang

Roman Igorevich Chernyak

Xianguo Zhang

Victor Alexeevich Stepin

Sijia Chen

Jian-Liang Lin

Yung-Chang Chang

Alexey Konstantinovich Filippov

Vasily Alexeevich Rufitskiy

Weijia Zhu

Yi Li

Lei Li

Ying Hu

Chih-Ming Fu

Sriram Sethuraman

Qing Zhang

Fai Yeung

Alexander Alexandrovich Karabutov

Chia-Yang Tsai

Arianne Hinds

Tao Zhang

Jeeva Raj A

Sagar Kotecha

A video bitstream including a current transform block (TB) in a current picture is received. A context model is determined for a syntax element associated with a transform coefficient level of a first coefficient group (CG) in the current TB based on transform coefficient levels of at least one first neighboring CG of the first CG. The first CG is positioned on a first scanning line. The at least one first neighboring CG is positioned on a second scanning line that is scanned before the first scanning line. The context model is a probability model for a non-separable transform. The first CG is reconstructed based on the transform coefficient level that is determined according to the determined the context model.

Context derivation for arithmetic coding of transform coefficients generated by non-separable transforms

Aspects of the disclosure provide methods and apparatuses for audio processing. In some examples, an apparatus of audio coding includes processing circuitry. The processing circuitry decodes, from a coded bitstream, information indicative of an adjusted speech signal and a loudness adjustment to the adjusted speech signal. The adjusted speech signal is indicated in an association with multiple speech signals in a scene of an immersive media application. The processing circuitry determines a plurality of loudness adjustments to sound signals including the multiple speech signals in the scene based the plurality of loudness adjustment to the adjusted speech signal, and generates the sound signals in the scene based on the loudness adjustments to the sound signals.

Method and apparatus in audio processing

Aspects of the disclosure provide methods and apparatuses for mesh coding (e.g., compression and decompression). In some examples, an apparatus for mesh coding includes processing circuitry. The processing circuitry decodes, using a point cloud compression (PCC) decoder and from a bitstream, a point cloud that includes points corresponding to at least one of vertices in a mesh or sampling points of polygons in the mesh. The processing circuitry determines, based on the point cloud, estimated connectivity information of the vertices in the mesh, and generate, a reconstructed mesh frame based on the point cloud and the estimated connectivity information.

Methods and apparatuses for dynamic mesh compression

The various embodiments described herein include methods and systems for encoding and decoding video. In one aspect, a method includes receiving video data comprising a plurality of blocks, including a first block. The method also includes obtaining, from the video data, a syntax element value indicating a wedge mode index corresponding to a first partition mode from a plurality of partition modes for the first block, where the wedge mode index is greater than 16. The method further includes determining that the first partition mode is associated with a compound wedge-based prediction for the first block. The method also includes partitioning the first block into a first section and a second section in accordance with the first partition mode and the wedge mode index, and decoding the first section in accordance with a first predictor and decoding the second section in accordance with a second predictor.

Systems and methods for wedge-based prediction mode signaling

A method of video decoding performed in a video decoder is provided. In the method, coded information of a current block is received from a coded video bitstream. The coded information includes a context-coded syntax element associated with the current block. The syntax element indicates a decoding parameter of the current block. A context-based adaptive binary arithmetic coding (CABAC) context model associated with the context-coded syntax element is determined based on decoding parameters of more than two spatially neighboring blocks of the current block. The context-coded syntax element is decoded based on the determined CABAC context model associated with the context-coded syntax element. The current block is reconstructed based on the decoded syntax element.

CABAC context modeling with information from spatial neighbors

Various implementations described herein include methods and systems for coding video. In one aspect, a video bitstream includes a current image frame and a first syntax element. The first syntax element has a first predefined value that indicates that the band offset only mode is enabled to determine a first sample offset of a first color sample based on luma samples, independently of any associated luma gradient of the luma samples. When the band offset only mode is enabled, quantized values are generated based on the luma samples including a first luma sample collocated with the first color sample. A first color sample is classified based on the quantized values to determine the first sample offset of the first color sample. The current image frame is reconstructed at least by adjusting the first color sample based on the first sample offset.

CCSO with band offset only option

A method performed in a decoder includes receiving a coded video bitstream that includes one or more 2D meshes corresponding to a 3D mesh. The method includes reconstructing one or more boundary vertices of each of the one or more 2D meshes. The method includes determining whether a reconstruction mode for reconstructing one or more internal vertices of each of the one or more 2D meshes is one of (i) a sampling mode and a (ii) non-sampling mode. The method includes reconstructing the one or more internal vertices of each of the one or more 2D meshes based on the determined reconstruction mode. The method further includes reconstructing the 3D mesh based on the reconstructed one or more boundary vertices of each of the one or more 2D meshes and the reconstructed one or more internal vertices of each of the one or more 2D meshes.

Chart based mesh compression

A video bitstream including coding information of a current block in a current picture is received, where the coding information indicates that the current block is coded with an intra template matching prediction mode. At least one template for predicting the current block is determined. Each of the at least one template is a respective candidate template of a plurality of template types available in the intra template matching prediction mode. The candidate templates are associated with one of a shape of the current block, a size of the current block, an intra prediction mode of the current block, and distances between the current block and the candidate templates of the plurality of template types. The candidate templates include different combinations of neighboring reconstructed samples. The current block is reconstructed using template match in the intra template matching prediction mode based on the determined at least one template.

Multi-template based intra-frame template matching prediction

Aspects of the disclosure provide a method and an apparatus including processing circuitry for video decoding. The processing circuitry can be configured to determine a transform skip for a transform block (TB) is in one direction. The processing circuitry can be configured to quantize transform coefficients in the TB based on a plurality of quantization steps. First transform coefficients in the TB can be quantized based on a smallest of the plurality of quantization steps. The first transform coefficients in the TB include one of a first row and a first column of the transform coefficients in the TB that has a DC spatial frequency, the one of the first row and the first column being along the one direction of the transform skip. Further, the processing circuitry can be configured to perform a transform on the quantized transform coefficients in the TB.

Quantizer for one-dimensional transform skip

An example method includes receiving a video bitstream comprising a plurality of coding blocks. The method includes identifying, based on a first indicator in the video bitstream, a coding region that comprises two or more luma blocks and two or more chroma blocks. A subset of luma blocks in the coding region is encoded in a first prediction mode. Each chroma block in the coding region is encoded in a second prediction mode. The method includes applying a first partitioning to the two or more luma blocks in the coding region. The method includes applying a second partitioning for the two or more chroma blocks in the coding region. The method includes reconstructing the coding region using the first partitioning and the second partitioning.

Systems and methods for recursive inter region partitioning

Processing circuitry decodes, from a bitstream carrying a 3D mesh, coordinates of vertices in the 3D mesh. The 3D mesh represents a surface of an object with polygons, the polygons are defined by the vertices and an original connectivity having original edges connecting the vertices. The processing circuitry derives an inferred connectivity according to the coordinates of the vertices, the inferred connectivity has inferred edges connecting the vertices. The processing circuitry decodes, from the bitstream, connectivity differences. In an example, the connectivity differences are associated with a subset of the vertices. In another example, the connectivity differences are associated with one or more edges in the original edges and the inferred edges. The processing circuitry determines a recovered connectivity according to the inferred connectivity and the connectivity differences, and reconstructs the 3D mesh according to the recovered connectivity.

Vertex connectivity coding for mesh compression

Methods and apparatuses for performing cross-component intra prediction, including: receiving a current chroma block from a coded bitstream; determining, from the coded bitstream, that a scaling factor used for the current chroma block in a chroma for luma (CfL) intra prediction mode; deriving a predicted scaling factor based on first neighboring samples of the current chroma block and second neighboring samples of a luma block that is co-located with the current chroma block; using the predicted scaling factor as the scaling factor used for the current chroma block in the CfL intra prediction mode; reconstructing the current chroma block after scaling the current chroma block based on the predicted scaling factor.

Signaling method for scaling parameter in chroma from luma intra prediction mode

A method, computer program, and computer system is provided for coding video data. Video data including one or more blocks is received. A current block coded in intra block copy mode or string matching mode is predicted from among the one or more blocks based on a coded block vector or a string offset vector corresponding to one or more spatial neighboring or non-neighboring blocks from among the one or more blocks. The video data is decoded based on the predicted current block.

Spatial displacement vector prediction for intra picture block and string copying

Content-adaptive online training for end-to-end (E2E) neural image compression (NIC) using a neural network performed by at least one processor, is provided, including receiving an input image, to an E2E NIC framework, splitting the input image into a plurality of blocks, selecting a subset of blocks from the plurality of blocks, the subset of blocks sharing a same pattern, preprocessing a neural network of the E2E NIC framework, wherein the preprocessed neural network is applied to the selected subset of blocks, computing updated parameters using the preprocessed neural network, and generating an updated E2E NIC framework, based on the updated parameters.

System, method, and computer program for content adaptive online training for multiple blocks based on certain patterns

The various embodiments described herein include methods and systems for coding video. In one aspect, a method includes obtaining video data comprising a plurality of blocks, including a first block. The method further includes determining a plurality of transform coefficients associated with the first block; and selecting a subset of the plurality of transform coefficients based on a frequency corresponding to each transform coefficient of the plurality of transform coefficients. The method also includes predicting respective coefficient signs for selected subset of the transform coefficients associated with the first block based on respective context information; and reconstructing the first block based on the plurality of transform coefficients and the predicted respective coefficient signs.

Systems and methods for frequency-dependent coefficient sign coding

Growing community of inventors

San Jose, CA, United States of America

Shan Liu