Video Frame Interpolation with Picture Switch Detection
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current video frame interpolation algorithms face challenges in handling deformation issues caused by scene and subtitle switches, leading to obvious deformations during frame interpolation processing, which affects video smoothness and user experience.
Innovation Solution
A method for video frame interpolation that acquires adjacent video frames, determines if a picture switch exists between them, and decides whether to interpolate a frame based on the comparison result, using optical flow prediction and text recognition to accurately assess subtitle and scene changes, thereby avoiding deformation.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Stability of the object's composition
If frame interpolation is performed between all adjacent video frames, then video smoothness is improved, but deformation artifacts occur during scene and subtitle switches
Solution Approach 1:
The patent applies preliminary action by detecting picture switches (scene and subtitle switches) between adjacent video frames before performing frame interpolation. The system compares subtitle text content and detects scene changes using image processing techniques to identify frames where interpolation should be skipped, thereby preventing deformation artifacts from occurring in the first place
Solution Approach 2:
The patent applies local quality by selectively applying frame interpolation only to specific frame pairs where no picture switch is detected. Instead of uniformly interpolating all frames, the system identifies local regions (specific frame intervals) where interpolation is safe and applies it only there, while skipping frames where picture switches occur, thus maintaining high quality overall
2Manufacturing precision
If frame interpolation is skipped during picture switches, then deformation artifacts are avoided, but video smoothness deteriorates
Solution Approach 1:
The patent introduces an intermediary mechanism - a picture switch detection module that acts as a mediator between the original video frames and the frame interpolation process. This intermediary detects the presence of picture switches and controls whether interpolation should be applied, thereby resolving the conflict between maintaining accuracy and ensuring smoothness by making intelligent decisions based on content analysis
3Measurement precision
If picture switch detection is performed using text recognition, then subtitle switch accuracy is improved, but processing complexity increases
Solution Approach 1:
The patent replaces traditional mechanical or manual subtitle detection methods with text recognition technology (such as OCR or AI-based text extraction). This substitution enables automatic, accurate detection of subtitle changes by comparing recognized text content between frames, significantly improving detection accuracy while the automated nature actually reduces operational complexity despite increasing computational requirements
Data Source
AI summary
A method and an apparatus for video frame interpolation processing, and a non-instantaneous readable storage medium. The method for video frame interpolation processing, including: acquiring the first video frame and the second video frame of the video; acquiring the first comparison result between the first video frame and the second video frame based on the first video frame and the second video frame; and determining whether to interpolate a frame between the first video frame and the second video frame based on the first comparison result. The first video frame and the second video frame are adjacent temporally, and the first video frame is a forward frame of the second video frame. The first comparison result indicates whether a picture switch exists between the first video frame and the second video frame.