Trim-pass correction for cloud-based HDR video coding
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
In cloud-based video coding of HDR video, existing techniques face challenges in efficiently managing trim-pass correction, leading to unacceptable overhead and temporal inconsistencies across computing nodes.
Innovation Solution
The proposed solution involves generating scene-based trim-pass correction parameters and forward reshaping functions in a cloud-based system, where each node processes a video sequence and adjusts trim-pass parameters based on frame-based and scene-based data to maintain temporal consistency.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Manufacturing precision
If frame-based trim-pass correction parameters are generated for each video frame, then coding precision is improved, but device complexity and computational overhead increase significantly
Solution Approach 1:
The patent segments the video sequence into scenes and further divides scenes into sub-scenes, generating trim-pass correction parameters at the sub-subscene level rather than for every frame. This hierarchical segmentation reduces the total number of parameter generations while maintaining visual consistency within each segment, resolving the contradiction between coding precision and device complexity.
Solution Approach 2:
The patent applies trim-pass correction parameters selectively only to sub-scenes where visual inconsistencies are detected, rather than uniformly to all frames. This partial action approach maintains coding precision where needed while reducing overall computational overhead and device complexity.
2Productivity
If cloud-based coding processes video sequences distributed across multiple computing nodes, then productivity is improved, but temporal inconsistencies arise across nodes
Solution Approach 1:
The patent performs preliminary scene segmentation and identifies scene boundaries before distributing video sequences to multiple computing nodes. By pre-establishing scene structure and generating trim-pass correction parameters at scene boundaries, the system ensures temporal consistency is maintained across nodes while allowing parallel processing within each node, thus resolving the contradiction between productivity and temporal stability.
3Adaptability or versatility
If HDR video is mapped to SDR format for display compatibility, then adaptability is improved, but visual quality deteriorates due to artifacts
Solution Approach 1:
The patent dynamically adjusts trim-pass correction parameters based on local scene characteristics, luminance levels, and contrast properties during the HDR to SDR mapping process. By changing these parameters adaptively rather than using fixed mappings, the system maintains visual quality and reduces artifacts while ensuring display compatibility, thus resolving the contradiction between adaptability and manufacturing precision.
Data Source
AI summary
In a cloud-based system for encoding high dynamic range (HDR) video, each node receives a video segment and bumper frames. Each segment is subdivided into primary scenes and secondary scenes to derive scene-based forward reshaping functions that minimize the amount of reshaping-related metadata when coding the video segment. When a parent scene of a secondary scene is processed by two or more neighboring nodes, initial forward reshaping functions and trim-pass correction parameters are adjusted using reference tone-mapping functions and updated scene-based trim-pass correction parameters.