Video Encoding Mode Class Selection via Statistical Properties
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current digital video encoding methods face high computational complexity in determining the optimal prediction mode for video blocks, making real-time processing challenging due to the need to evaluate multiple prediction modes and compute prediction errors.
Innovation Solution
The method reduces computational complexity by selecting a mode class based on statistical properties of the video information, such as pixel intensity variations, eliminating the need to generate or evaluate prediction errors for all modes, and using statistical learning to differentiate between mode classes.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Manufacturing precision
If multiple prediction modes are evaluated to determine optimal encoding mode, then encoding performance is improved, but computational complexity increases
Solution Approach 1:
The patent segments the mode selection process into two stages: first selecting a mode class based on statistical properties, then selecting a specific mode within that class. This segmentation reduces the overall complexity by dividing the search space into manageable parts.
Solution Approach 2:
The patent performs preliminary classification of macroblocks into mode classes based on statistical properties before conducting the actual mode selection. This preliminary action eliminates the need to evaluate all modes for all macroblocks, significantly reducing computational complexity.
2Manufacturing precision
If all prediction modes are searched to find best compression algorithm, then compression performance is improved, but processing time increases
Solution Approach 1:
The patent applies partial action by evaluating only a subset of modes (those within the selected mode class) rather than all possible modes. This partial evaluation achieves satisfactory compression performance while significantly reducing processing time.
Solution Approach 2:
The patent changes the parameter being optimized from direct compression performance evaluation to statistical property-based classification. This parameter change enables faster mode class selection while maintaining acceptable compression performance.
3Measurement precision
If prediction error computation is performed for mode selection, then mode accuracy is improved, but computational complexity increases
Solution Approach 1:
The patent extracts and uses only the statistical properties of macroblocks (such as variance, gradient, or texture features) for mode class selection, taking out this specific information from the full macroblock data. This extraction avoids the need to compute prediction errors while still enabling effective mode selection.
Solution Approach 2:
The patent introduces statistical properties as an intermediary between the macroblock content and the mode selection process. Instead of directly computing prediction errors, the statistical properties serve as a mediator to guide mode class selection, reducing computational complexity.
Data Source
AI summary
An improvement to a predictive video encoding method or apparatus that includes selecting a mode class within which to choose an encoding algorithm to encode individual blocks of video information. Rather than implementing each algorithm in each mode class to ascertain an acceptable compression, the improvement eliminates searching through a class of encoding modes based simply on heuristics. The method comprises obtaining statistical information related to previous blocks of encoded video information, determining a mode class within which to chose a particular encoding algorithm based on the statistical information (e.g., heuristics) gathered, choosing an algorithm within the selected mode class using conventional techniques, and encoding the video information according to the chosen algorithm. Statistical information may include quantization parameters, prior encoding decisions, intensity or frequency values, or Hadamard transform coefficients of previously encoded macroblocks. Encoding complexity in the coder is reduced since one class of encoding modes is eliminated.