Video Decoder Joint Transform Type and Subblock Pattern VLC
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current video decoding techniques, such as those in WMV8, face inefficiencies in signaling overhead and performance due to independent signaling of transform sizes, types, and subblock patterns, especially when every block in a frame has a specified transform size, and the use of small event spaces for transform type and subblock pattern signaling.
Innovation Solution
A video decoder that jointly encodes and decodes transform type and subblock pattern information using variable length codes (VLCs), allowing for improved performance by selecting appropriate VLC tables and signaling mechanisms that combine transform signal level, type, and subblock pattern information at different levels (macroblock and block levels).
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Ease of manufacture
If transform type and subblock pattern are signaled independently using separate VLC tables, then the signaling mechanism is simple and easy to implement, but the signaling overhead increases and coding efficiency decreases
Solution Approach 1:
The patent combines transform type signaling and subblock pattern signaling into a single joint VLC table. Instead of using separate VLC tables for transform type and subblock pattern as in prior art, this invention creates unified VLC tables that encode both pieces of information together, reducing the total number of bits required for signaling while maintaining implementation feasibility.
2Device complexity
If separate VLC tables are used for transform type and subblock pattern, then the coding structure is simple and modular, but the event space remains limited and coding efficiency is reduced
Solution Approach 1:
The invention merges transform type and subblock pattern into a single joint VLC table, creating an expanded event space that combines both parameters. This allows the decoder to efficiently determine both transform type and subblock pattern from a single coded value, improving coding efficiency without significantly increasing structural complexity.
Data Source
AI summary
Techniques and tools are described for decoding jointly information. For example, a decoder decodes a variable length [“VLC”] signaled at macroblock level that jointly represents a transform type signal level, transform type, and subblock pattern. The decoder decodes one or more VLCs signaled at block level, each jointly representing a transform type and subblock pattern. The decoder may select between multiple VLC tables for the VLCs signaled macroblock level and/or block level.