Transform Unit Size Determination via Activity Computation
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
The existing video encoding systems, such as those using HEVC, face increased processing load due to the need for orthogonal transformation across all TU sizes to determine the optimum size for improved encoding efficiency.
Innovation Solution
An image encoding apparatus that determines the Transform Unit (TU) size based on activity or motion information within a Coding Unit (CU), optimizing processing units such as Coding Units, Prediction Units, and Transform Units by dividing images into blocks and computing activities or motion information to set the TU size efficiently, thereby reducing processing load.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Productivity
If orthogonal transformation is performed for all TU sizes to determine the optimum TU size, then encoding efficiency is improved, but encoding processing load increases
Solution Approach 1:
The patent changes the parameter used for TU size determination from RD cost (which requires full orthogonal transformation) to activity values and motion information. By using these alternative parameters, the system can determine optimal TU sizes without performing computationally intensive orthogonal transformations for all candidate sizes, thus reducing encoding processing load while maintaining encoding efficiency
Solution Approach 2:
The patent extracts only the necessary information (activity values and motion information) needed for TU size determination, rather than performing complete orthogonal transformations. This extraction approach allows the system to obtain sufficient data for decision-making while avoiding unnecessary computational steps, thereby reducing the encoding processing load
2Productivity
If multiple TU sizes are evaluated to optimize processing units, then encoding efficiency is improved, but device complexity increases
Solution Approach 1:
The patent simplifies the evaluation process by changing from comprehensive RD cost calculation to using activity values and motion information as determination criteria. This parameter substitution reduces the computational steps required for evaluating multiple TU sizes, thereby reducing device complexity while still enabling optimization of processing units
Solution Approach 2:
The patent segments the TU size determination process into distinct evaluation stages based on activity values and motion information thresholds. By dividing the evaluation into manageable segments rather than performing exhaustive comparisons, the system reduces processing complexity while maintaining the ability to optimize encoding efficiency
Data Source
AI summary
An image encoding method that encodes an image so as to optimize estimated values of processing units which are a coding unit, a prediction unit, and a transform unit by selecting sizes of the individual processing units. The method includes a division step that divides an image having a size of the coding unit into a plurality of blocks; an activity computation step that computes activities pertaining to the individual blocks; and a transform unit size determination step that determines, based on the computed activities, a size pattern of the transform unit within the image having the size of the coding unit.