Invention Application
- Patent Title: ADAPTIVE CODING OF PREDICTION MODES USING PROBABILITY DISTRIBUTIONS
-
Application No.: US17340293Application Date: 2021-06-07
-
Publication No.: US20220014744A1Publication Date: 2022-01-13
- Inventor: Yaowu Xu , Paul Gordon Wilkins , James Bankoski
- Applicant: GOOGLE LLC
- Applicant Address: US CA Mountain View
- Assignee: GOOGLE LLC
- Current Assignee: GOOGLE LLC
- Current Assignee Address: US CA Mountain View
- Main IPC: H04N19/122
- IPC: H04N19/122 ; H04N19/13

Abstract:
Generating, by a processor in response to instructions stored on a non-transitory computer readable medium, a reconstructed frame, may include generating a reconstructed block of the reconstructed frame by decoding from an encoded bitstream. Decoding may include decoding a value from the encoded bitstream, identifying, in accordance with the value, a probability distribution for generating the reconstructed block, wherein the value indicates the probability distribution among a plurality of probability distributions determined independently of generating the reconstructed frame, entropy decoding an encoded prediction mode from the encoded bitstream using the probability distribution to identify a prediction mode for generating the reconstructed block, generating a prediction block in accordance with the prediction mode; combining the prediction block and a reconstructed residual block to obtain the reconstructed block, and including the reconstructed block in the reconstructed frame.
Public/Granted literature
- US11627321B2 Adaptive coding of prediction modes using probability distributions Public/Granted day:2023-04-11
Information query