公开(公告)号：US20250119577A1

公开(公告)日：2025-04-10

申请号：US18901214

申请日：2024-09-30

Applicant: GOOGLE LLC

Inventor： Xiang Li ,  Debargha Mukherjee ,  Yaowu Xu ,  Jingning Han

IPC: H04N19/593 ,  H04N19/11 ,  H04N19/117 ,  H04N19/119 ,  H04N19/136 ,  H04N19/176 ,  H04N19/184 ,  H04N19/186

Abstract: Encoding using chroma intra prediction with filtering includes encoding a current block from a current frame, which includes obtaining a first chroma prediction value for a current chroma pixel using a current spatial intra prediction mode, obtaining a current luma prediction value for a current luma pixel collocated with the current chroma pixel, obtaining a second chroma prediction value for the current chroma pixel for the current chroma component by applying derived filter coefficients to the current luma prediction value, obtaining, as a third chroma prediction value for the current chroma pixel for the current chroma component, a weighted average of the first chroma prediction value and the second chroma prediction value, obtaining encoded chroma pixel data for the current chroma pixel by encoding the current chroma pixel using the third chroma prediction value, and including the encoded chroma pixel data in the encoded block data.

公开(公告)号：US20250039436A1

公开(公告)日：2025-01-30

申请号：US18785699

申请日：2024-07-26

Applicant: GOOGLE LLC

Inventor： Lester Lu ,  Xiang Li ,  Debargha Mukherjee

IPC: H04N19/513 ,  H04N19/139 ,  H04N19/176 ,  H04N19/196

Abstract: Coding including dynamic range handling of high dimensional inverse autocorrelation in optical flow refinement includes obtaining a refinement model from available warped refinement models, wherein the available warped refinement models include a four-parameter scaling refinement model, a three-parameter scaling refinement model, and a four-parameter rotational refinement model, obtaining refined motion vectors using the warped refinement model and previously obtained reference frame data in the absence of data expressly indicating the refined motion vectors in the encoded bitstream, wherein obtaining the refined motion vectors includes using a dynamic range adjusted autocorrelation matrix, generating refined prediction block data using the refined motion vectors, generating reconstructed block data using the refined prediction block data, including the reconstructed block data in reconstructed frame data for the current frame, and outputting the reconstructed frame data.

公开(公告)号：US20240223796A1

公开(公告)日：2024-07-04

申请号：US18390238

申请日：2023-12-20

Applicant: GOOGLE LLC

Inventor： Debargha Mukherjee ,  Mohammed Golam Sarwer ,  Rachel Barker ,  Jianle Chen ,  Xiang Li

IPC: H04N19/52 ,  H04N19/109 ,  H04N19/139 ,  H04N19/17 ,  H04N19/176 ,  H04N19/184

CPC classification number: H04N19/52 ,  H04N19/109 ,  H04N19/139 ,  H04N19/17 ,  H04N19/176 ,  H04N19/184

Abstract: Coding using local global prediction modes with projected motion fields includes identifying a current frame, identifying a current reference frame, obtaining a projected motion field, for the current frame, using motion data from the current reference frame, identifying a current superblock from the current frame, obtaining reference warp motion parameters for the current superblock by fitting the projected motion field to a warp motion model, and using the reference warp motion parameters to code respective blocks from the superblock.