On Screen Display Edge Detection for Video Interpolation
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Video frame interpolation in situations involving on-screen displays (OSDs) often results in artifacts such as broken OSDs and dirty backgrounds due to incorrect motion vectors, as OSDs are semi-transparent and can move during frame rate conversion.
Innovation Solution
A method is implemented to detect the edges of OSDs and assign a zero motion vector (0MV) to OSD blocks, using confidence values to accurately identify and isolate OSD regions within the video frames, thereby preventing artifacts during interpolation.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Speed
If video frame interpolation is performed using motion vectors for all regions, then the smoothness and clarity of moving video content is improved, but artifacts such as broken OSDs and dirty backgrounds appear in OSD regions
Solution Approach 1:
The video frame is divided into different regions: OSD regions and non-OSD regions. Motion vectors are calculated and applied differently for each region type. This segmentation allows the system to maintain high frame interpolation smoothness in video content regions while preventing artifacts in OSD regions by using zero motion vectors there.
Solution Approach 2:
Different quality settings are applied to different parts of the image. OSD regions use zero motion vectors to maintain sharp, artifact-free display, while non-OSD regions use calculated motion vectors to achieve smooth interpolation. This local quality approach ensures each region receives the appropriate processing for its specific requirements.
2Reliability
If zero motion vector is used for OSD regions to prevent artifacts, then OSD display accuracy is improved, but the overall frame interpolation smoothness is reduced
Solution Approach 1:
The processing is segmented so that only OSD regions use zero motion vectors while the majority of the frame (non-OSD regions) uses calculated motion vectors for smooth interpolation. This minimizes the impact on overall frame interpolation smoothness while still achieving accurate OSD display where needed.
Solution Approach 2:
The system applies different processing qualities locally: high-quality motion-compensated interpolation for video content regions and zero-motion-vector processing for OSD regions. This ensures that the reduction in interpolation smoothness is localized only to the small OSD areas rather than affecting the entire frame.
3Measurement precision
If motion vectors are calculated for all pixel blocks, then the precision of motion compensation is improved, but the computational complexity and processing time increase
Solution Approach 1:
The processing workload is segmented by identifying and separating OSD regions from non-OSD regions. Motion vectors are calculated only for non-OSD regions, while OSD regions are processed with zero motion vectors. This segmentation significantly reduces the number of motion vector calculations required, lowering computational complexity while maintaining accuracy where it matters.
Solution Approach 2:
Instead of calculating motion vectors for all blocks (excessive action), the system calculates motion vectors only for non-OSD blocks where they are actually needed (partial action). This partial processing approach reduces computational complexity while maintaining sufficient motion compensation accuracy for the regions that require it.
Data Source
AI summary
An on-screen display (OSD) processing system can include video data fetch modules to receive video data from a video source, edge detection modules to detect an edge of a region within a scene represented by the video data, edge processing modules to process results from the edge detection modules, and an OSD confidence level adjustment module to adjust an OSD confidence level corresponding to the region within the scene.