Low Dynamic Range Signaling in Video Coding Bitstreams
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing video coding technologies face challenges in efficiently signaling Low Dynamic Range (LDR) information in video or image bitstreams, which affects coding efficiency and accuracy.
Innovation Solution
The proposed solution involves methods and systems for explicitly or implicitly signaling whether LDR is applied to a block and which LDR is used, using predefined LDRs and their corresponding indexes signaled through various syntax elements in the bitstream.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If LDR information is explicitly signaled in the bitstream using syntax elements, then coding precision and control over LDR application is improved, but signaling overhead and bitstream complexity increases
Solution Approach 1:
The patent implements dynamic LDR signaling by introducing syntax elements that allow the LDR applied flag and LDR index to be signaled at different hierarchical levels (sequence, picture, slice, or block level). This dynamic approach enables adaptive LDR application based on local image characteristics and coding conditions, improving coding precision while allowing flexibility to minimize signaling overhead by selecting the appropriate signaling level.
Solution Approach 2:
The patent applies LDR signaling at different granularities including block-level syntax elements, allowing different LDR parameters to be applied to different blocks within the same picture. This local quality approach enables precise control over LDR application in specific regions where it is most beneficial, while avoiding unnecessary signaling overhead in regions where LDR is not applied or where higher-level signaling suffices.
2Productivity
If LDR is applied to reduce bit depth of video samples, then coding efficiency and compression ratio are improved, but reconstruction accuracy and image quality may deteriorate
Solution Approach 1:
The patent performs LDR mapping as a preliminary action during the encoding process, transforming high-bit-depth samples to lower bit-depth LDR samples using predefined transfer functions before compression. This preliminary transformation allows the subsequent compression stages to work with reduced data precision, improving coding efficiency while the decoder can apply inverse mapping to reconstruct the image, balancing compression ratio and reconstruction accuracy.
Solution Approach 2:
The patent changes the bit depth parameter of video samples by applying LDR mapping with predefined ranges (e.g., 10-bit to 8-bit, 8-bit to 6-bit). This parameter change reduces the dynamic range of the signal, allowing more aggressive compression with fewer bits per sample, thereby improving coding efficiency while maintaining acceptable reconstruction accuracy through the structured LDR framework.
3Adaptability or versatility
If multiple LDR options are provided with indexes signaled in the bitstream, then adaptability and versatility of LDR application are improved, but device complexity and processing overhead increase
Solution Approach 1:
The patent provides multiple LDR options with different bit depth reductions (e.g., LDR index 0 for 10-bit to 8-bit, index 1 for 8-bit to 6-bit) that can be dynamically selected and signaled in the bitstream. This dynamic selection mechanism improves adaptability by allowing the encoder to choose the most appropriate LDR level based on content characteristics and coding conditions, while the decoder complexity is managed through predefined LDR transfer functions that simplify the inversion process.
Data Source
AI summary
This disclosure relates generally to video coding and particularly to methods and systems for signaling of Low Dynamic Range (LDR) in video or image bitstream. For example, the disclosure describes implementations of manners for signaling whether LDR is applied and which one of LDRs is applied to a block at various singling levels, either explicitly or implicitly.