Video Encoder Intra-Picture Prediction Mode Search Strategy
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current video encoding processes are computationally intensive due to the evaluation of numerous intra-picture prediction modes (IPPMs), particularly in scenarios demanding high temporal and spatial resolution, leading to time-consuming and resource-intensive operations.
Innovation Solution
Implementing strategic search methods for a video encoder to selectively skip the evaluation of certain IPPMs that are unlikely to improve rate-distortion performance, such as using gradient searches only when likely to improve performance, comparing encoding costs with thresholds, and prioritizing evaluations based on angular modes and chroma sample values.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If exhaustive evaluation of multiple intra-picture prediction modes is performed, then rate-distortion performance is improved, but encoding time and computational complexity increase significantly
Solution Approach 1:
The patent applies partial action by selectively evaluating only a subset of intra-picture prediction modes based on gradient calculations. Instead of exhaustively evaluating all possible modes, the encoder performs gradient searches only when the gradient magnitude exceeds a threshold, thereby performing partial evaluation that suffices for achieving good rate-distortion performance without the full computational burden of exhaustive search
Solution Approach 2:
The patent changes the parameter of evaluation thoroughness dynamically based on gradient magnitude. When gradient magnitude is below a threshold, the encoder skips detailed mode evaluation; when above the threshold, it performs gradient search. This parameter-based conditional evaluation resolves the contradiction by adapting the evaluation depth to the actual need, reducing encoding time while maintaining performance
2Reliability
If gradient search among angular modes is performed to find suitable prediction modes, then rate-distortion performance improves, but computational complexity increases
Solution Approach 1:
The patent uses gradient magnitude as a parameter to control whether gradient search is performed. By comparing the gradient magnitude against a threshold, the system dynamically changes the computational approach: skipping gradient search when gradient is small (low complexity) and performing it only when gradient is large (high complexity only when needed). This resolves the contradiction by making computational complexity adaptive rather than constant
Solution Approach 2:
The patent applies skipping by bypassing the gradient search process entirely when the gradient magnitude threshold is not met. This allows the encoder to rush through difficult computational steps when they are not necessary, achieving good rate-distortion performance through simpler means and avoiding the computational complexity of gradient search only when truly needed
3Reliability
If comprehensive evaluation of all intra-picture prediction modes is conducted, then encoding quality is improved, but resource consumption increases
Solution Approach 1:
The patent applies partial action by evaluating only the necessary subset of prediction modes. Through gradient-based early termination and selective gradient search, the system performs just enough evaluation to achieve good encoding quality without conducting exhaustive searches, thereby reducing computational resource consumption while maintaining quality
Solution Approach 2:
The patent changes the evaluation depth parameter based on gradient magnitude and threshold comparisons. When gradients are small, evaluation is terminated early; when gradients are large, more thorough evaluation is performed. This dynamic parameter adjustment ensures encoding quality is maintained while adapting resource consumption to actual content characteristics
Data Source
AI summary
Innovations are presented that reduce the computational complexity of video encoding by selectively skipping certain evaluation stages during intra-picture prediction. A video encoder receives and encodes a current picture. As part of the encoding, for a current block of the current picture, the video encoder evaluates at least some intra-picture prediction modes (“IPPMs”). According to a search strategy, the video encoder selectively skips time-consuming evaluation of certain IPPMs for the current block when those IPPMs are not expected to improve rate-distortion performance, which can dramatically speed up the encoding process. For example, the video encoder conditionally performs a gradient search among angular IPPMs. Or, as another example, the video encoder selectively skips evaluation of IPPMs depending on a cost of encoding the current block using motion compensation. Or, as another example, the video encoder prioritizes IPPMs evaluated for a block of chroma sample values.