Video Processing Illuminance-Based Optical Flow Coding
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional video coding techniques have low coding efficiency, and existing optical flow based coding methods do not effectively handle illuminance changes, which affects coding performance.
Innovation Solution
A method that determines whether to apply optical flow based coding to a video unit based on illuminance information, adjusting the coding process by considering illuminance changes and using tools like local illumination compensation and bi-prediction with CU-level weight to refine predictions.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If optical flow based coding method is applied to all video units, then coding performance is improved, but coding complexity increases
Solution Approach 1:
The patent changes the parameter of coding method selection based on illuminance information. By calculating illuminance change between current and reference video units, the system dynamically adjusts whether to apply optical flow based coding, thereby optimizing performance while controlling complexity through conditional application rather than universal application
Solution Approach 2:
The patent applies different coding qualities to different regions based on illuminance characteristics. By identifying video units with significant illuminance changes and applying optical flow based coding selectively to those regions, the system achieves local optimization where it is most needed while maintaining lower complexity in stable regions
2Device complexity
If conventional video coding techniques are used, then device complexity is reduced, but coding efficiency deteriorates
Solution Approach 1:
The patent introduces dynamic adaptability to video coding by using illuminance information to determine coding method selection. The system transitions from static conventional coding to dynamic conditional coding, where the coding approach adapts based on scene characteristics, thereby improving efficiency without requiring permanently complex device architecture
Solution Approach 2:
The patent segments the video content into different types based on illuminance characteristics. By dividing video units into those with significant illuminance changes and those without, the system applies appropriate coding strategies to each segment, improving overall efficiency while keeping individual processing tasks manageable
3Speed
If optical flow based coding is applied without considering illuminance information, then coding speed is maintained, but coding accuracy deteriorates
Solution Approach 1:
The patent performs preliminary calculation of illuminance information before applying optical flow based coding. By pre-calculating illuminance changes and using this information to guide subsequent coding decisions, the system ensures accurate coding application while maintaining efficient processing through avoided unnecessary calculations
Data Source
AI summary
Embodiments of the present disclosure provide a solution for video processing is proposed. A method for video processing is proposed. The method comprises: determining, during a conversion between a video unit and a bitstream of the video unit, information on applying an optical flow based coding method to the video unit based on illuminance information of the video unit; and performing the conversion based on the information. Compared with the conventional solution, the proposed method can advantageously improve the coding efficiency and performance.