Video Pixel Prediction Using Intensity-Difference Patterns
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing video compression technologies struggle with accurately predicting pixel changes in regions with motion, leading to decreased video quality or increased data transmission requirements, especially in areas with differently shaded objects or sections.
Innovation Solution
The method involves detecting patterns of pixel intensity differences in neighboring reconstructed pixels to enhance intra- and inter-prediction processes, using spatial and temporal trends to improve prediction accuracy and reduce data requirements.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Loss of information
If conventional inter-prediction using motion vectors is used, then data transmission is reduced for static regions, but prediction accuracy deteriorates in regions with motion and shading changes
Solution Approach 1:
The patent applies different prediction strategies to different regions of the image based on local characteristics. For regions with motion or shading changes, it uses gradient-based prediction modes that adapt to local intensity variations, while other regions may use conventional motion compensation. This local adaptation resolves the contradiction by optimizing prediction accuracy for specific problematic regions without unnecessarily increasing data transmission for entire frames.
Solution Approach 2:
The patent changes the prediction parameters by introducing gradient calculations and multiple prediction modes that can be selected based on regional characteristics. Instead of using a single motion vector approach, it employs parameter variations including spatial gradients, temporal gradients, and different prediction directions to match local image characteristics, thereby improving prediction accuracy for regions with motion and shading changes.
2Measurement precision
If more data is transmitted to maintain video quality in high-frequency detail regions, then video quality is maintained, but data transmission requirements increase
Solution Approach 1:
The patent applies the partial action principle by using enhanced gradient-based prediction only where needed - specifically in regions with motion or shading changes - rather than applying it uniformly across the entire image. This selective application maintains video quality in problematic regions while avoiding unnecessary data transmission increases for regions that are adequately handled by conventional methods.
3Productivity
If conventional intra-prediction using spatially neighboring pixels is used, then encoding is efficient, but prediction accuracy deteriorates when reference pixels and current pixels belong to differently shaded areas
Solution Approach 1:
The patent introduces gradient calculations as an additional parameter set for intra-prediction. By computing spatial and temporal gradients, it creates prediction modes that adapt to local intensity variations and shading changes. This allows the system to maintain encoding efficiency while improving prediction accuracy in regions with shading differences through parameter variation rather than requiring more complex encoding structures.
Data Source
AI summary
Methods and apparatus encode image frames using intra-frame prediction by predicting pixels for a block of current pixels, based on a detected spatial pattern of pixel intensity differences among a plurality of neighboring reconstructed pixels to the block of current pixels, and encode a block of pixels of the image frame using the predicted block of reconstructed pixels. Inter-frame prediction is provided by determining whether blocks of pixels in temporally neighboring reconstructed frames corresponding to a candidate motion vector have a pattern of pixel intensity differences among the blocks from temporally neighboring frames. Predicted blocks are produced for a reconstructed frame based on the determined pattern of pixel intensity difference among temporally neighboring frames.