Lossless Mode Signaling in HEVC Video Coding Systems
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
The current High Efficiency Video Coding (HEVC) standard lacks efficient signaling mechanisms to indicate when an entire picture or slice is coded in lossless mode, leading to unnecessary inclusion of lossy coding-related information and processing, which can introduce distortion and reduce coding efficiency.
Innovation Solution
Incorporating new syntax elements such as cu_transquant_bypass_forced_flag and cu_transquant_bypass_forced_present_flag at the picture and slice levels to explicitly indicate whether all coding units should be coded in lossless mode, thereby skipping lossy coding-related processes like deblocking filter and sample adaptive offset processing.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Productivity
If lossy coding processes (deblocking filter, SAO) are applied to lossless coded pictures, then processing completeness is maintained, but coding efficiency deteriorates and distortion is introduced
Solution Approach 1:
The patent implements dynamic processing by making the application of deblocking filter and SAO adaptive based on the coding mode. When lossless coding is detected (cu_transquant_bypass_flag=1), these filters are dynamically disabled. When lossy coding is used, the filters are applied normally. This conditional processing resolves the contradiction by adapting the processing pipeline to the specific coding scenario.
Solution Approach 2:
The patent segments the processing pipeline into distinct paths for lossless and lossy coding. By introducing the cu_transquant_bypass_flag, the system creates separate processing branches: one path applies deblocking and SAO for lossy coding, while the other path skips these operations for lossless coding. This segmentation allows each path to be optimized for its specific purpose without interference.
2Adaptability or versatility
If lossy coding-related information is included in bitstream for all coding units, then compatibility is maintained, but bitrate efficiency deteriorates
Solution Approach 1:
The patent applies partial action by selectively including or excluding lossy coding-related information based on the coding mode. When lossless coding is used (cu_transquant_bypass_flag=1), the patent omits deblocking parameters, SAO parameters, and transform coefficient coding information. When lossy coding is used, these elements are included. This selective inclusion resolves the contradiction by providing only the necessary information for each coding scenario.
3Adaptability or versatility
If cu_transquant_bypass_flag is signaled for each coding unit, then coding mode flexibility is improved, but signaling overhead increases
Solution Approach 1:
The patent applies preliminary action by establishing the lossless/lossy coding mode at the slice header level before processing individual coding units. The cu_transquant_bypass_flag is set at the slice level, and all subsequent coding units within that slice inherit this setting. This preliminary determination eliminates the need to signal the flag for each individual coding unit, significantly reducing signaling overhead while maintaining the ability to switch between coding modes at the slice level.
Data Source
AI summary
A method of efficient lossless mode syntax signaling for a coding system supporting both lossless and lossy coding is disclosed. In one embodiment, a first syntax element is incorporated in a picture level to indicate whether to use a forced lossless coding mode if lossless coding is allowed for the current picture. If the first syntax indicates that the forced lossless coding mode is selected, all coding units in the slices of the picture are coded using lossless coding. If the first syntax indicates that the forced lossless coding mode is not selected, each coding unit in a slice of the picture can be coded according to a second syntax element indicating whether each coding unit is coded using lossless coding or not. In another embodiment, the forced lossless mode can be applied to coding units on a slice by slice basis.