Cross-Component Sample Offset Filtering for Reconstructed Video Quality
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing video coding and decoding techniques face challenges in improving the quality of recovered video data, particularly in bandwidth-limited communication channels, due to lossy processes and the need for dynamic filtering adjustments.
Innovation Solution
A cross-component based filtering system is introduced, which includes a filter that uses a sample classifier to classify samples based on intensity and generate offsets. This system operates within the in-loop filtering process of video coders and decoders, enhancing sample classification and filtering efficiency.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Loss of energy
If lossy decoding processes are applied to reduce bandwidth, then bandwidth consumption is reduced, but video data quality deteriorates
Solution Approach 1:
The patent introduces an in-loop filter as an intermediary component between the lossy decoder and the prediction algorithms. This filter processes the recovered video data to remove artifacts and improve quality before the data is used in prediction, thereby mediating between the bandwidth-reducing lossy process and the quality requirements of video coding without requiring additional transmission bandwidth
Solution Approach 2:
The filtering system is designed to be self-contained within the decoder, using only the decoded video data itself to perform filtering operations. The filter analyzes the recovered video data and applies appropriate filtering to improve quality, allowing the system to service its own quality improvement needs without external intervention or additional information from the encoder
2Manufacturing precision
If filtering algorithms are applied to improve recovered video data quality, then video quality is improved, but system complexity increases
Solution Approach 1:
The patent implements a dynamic filtering system that adapts its behavior based on the characteristics of the input video data. The filter adjusts its parameters and processing intensity according to the local features of the video content, allowing it to provide optimal filtering for different regions and scenarios without requiring a fixed complex configuration for all possible cases
Solution Approach 2:
The filtering process is divided into distinct functional stages and components, with each stage handling specific aspects of the filtering task. This segmentation allows the complex filtering operation to be broken down into manageable, independently optimized modules that can be processed sequentially, reducing overall system complexity while maintaining high video quality
3Manufacturing precision
If dynamic filtering adjustments are made to improve video quality, then filtering performance is improved, but processing time increases
Solution Approach 1:
The patent applies filtering selectively to only those regions of the video data that benefit most from filtering, rather than uniformly processing all data at maximum intensity. By identifying and focusing computational resources on areas with significant artifacts or quality issues, the system achieves high filtering performance where needed while minimizing unnecessary processing time in already-quality regions
Data Source
AI summary
A cross-component based filtering system is disclosed for video coders and decoders. The filtering system may include a filter having an input for a filter offset and an input for samples reconstructed from coded video data representing a native component of source video on which the filter operates. The offset may be generated at least in part from a sample classifier that classifies samples reconstructed from coded video data representing a color component of the source video orthogonal to the native component according to sample intensity.