Local Weighted Prediction Coefficients for Video Encoding Illumination
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing video encoding algorithms, such as H.264/AVC, are sensitive to global illumination changes and struggle to accurately assign weighted prediction coefficients, leading to suboptimal compression and quality in scenes with multiple light sources or changing lighting conditions.
Innovation Solution
A method that performs initial weighted prediction estimation for each block of a reference frame, clusters these estimates, and assigns each block of the target image to corresponding regions based on clustering, allowing for flexible and accurate assignment of weighted prediction coefficients to account for local illumination changes.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If global weighted prediction coefficients are used for the entire reference frame, then the encoding process is simple and fast, but the prediction accuracy deteriorates in scenes with multiple light sources or changing illumination conditions
Solution Approach 1:
The patent divides the reference frame into multiple regions (e.g., left half and right half, or multiple macro blocks) and estimates separate weighted prediction coefficients for each region. This segmentation allows each region to have optimized coefficients that match its specific illumination conditions, thereby improving prediction accuracy without requiring the entire frame to use a single set of coefficients.
Solution Approach 2:
The patent applies different weighted prediction coefficients to different regions of the reference frame based on their local illumination characteristics. By analyzing the target frame and identifying regions with different lighting conditions (e.g., indoor lighting vs. outdoor lighting), the system assigns locally optimized coefficients to each region, enabling accurate prediction while adapting to local quality variations.
2Productivity
If motion estimation is performed at block granularity to minimize residual, then compression efficiency improves, but the system becomes sensitive to global illumination changes and fails to maintain accuracy under varying lighting conditions
Solution Approach 1:
The patent changes the parameters of the prediction model by introducing region-specific weighted prediction coefficients alongside motion vectors. Instead of using a single set of coefficients for the entire frame, the system estimates and applies different coefficients (scaling factors and offset values) for each region, making the compression system robust to illumination changes while maintaining high compression efficiency.
3Adaptability or versatility
If a single weighted prediction coefficient set is applied to the entire reference frame, then the encoding process remains simple, but the ability to handle multiple light sources and illumination effects deteriorates
Solution Approach 1:
The patent segments the reference frame into multiple regions and estimates separate weighted prediction coefficients for each region. This allows the system to adapt to multiple light sources by assigning different coefficients to regions under different illumination conditions, such as indoor lighting regions versus outdoor lighting regions, thereby improving adaptability without excessive complexity.
Solution Approach 2:
The patent applies weighted prediction coefficients to only those regions where illumination changes are detected, rather than uniformly applying coefficients to the entire frame. This partial action approach allows the system to handle multiple light sources effectively while avoiding unnecessary processing in regions with uniform illumination, thus balancing adaptability with computational complexity.
Data Source
AI summary
A method, apparatus, and manufacture for encoding a video sequence is provided. Encoding the video sequence includes performing a weighted prediction estimation between a reference frame of the video sequence and a target frame of the video sequence. Performing the weighted prediction includes performing an initial weighted prediction estimation for each block of the reference frame. Next, blocks are clustered according to their initial weighted prediction estimates. Then, each block of the target image is assigned to a corresponding region based on the clustering. During the video encoding, weighted-prediction is employed for each block according to its corresponding region.