Video Frame Small-Object Region Determination
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current video processing technologies fail to accurately determine small-object regions in video frames, leading to missing pixels in interpolated frames, which affects the playback quality of high-definition and ultra-high-definition videos due to background motion vector replacement during interpolation.
Innovation Solution
A method and apparatus that divide a video frame into regions, determine global and interframe motion vectors, and perform filtering to identify small-object regions, ensuring accurate determination and preservation of these regions during interpolation between adjacent frames.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Stability of the object's composition
If smooth filtering is performed on motion vectors obtained after motion estimation, then the interpolated frame achieves desirable continuity with neighboring frames, but motion vectors corresponding to small-object regions are replaced by background motion vectors causing pixel absence
Solution Approach 1:
The patent segments the video frame into multiple regions including background regions and small-object regions. By identifying and separately processing motion vectors for different regions, the patent applies smooth filtering only to background regions while preserving motion vectors in small-object regions, thus maintaining continuity where appropriate while preventing pixel absence in small-object areas.
Solution Approach 2:
The patent applies different processing qualities to different parts of the frame. Background regions receive smooth filtering to ensure continuity, while small-object regions are preserved with their original motion vectors to prevent pixel absence. This local differentiation resolves the contradiction by applying the appropriate processing quality to each region.
2Productivity
If motion estimation is performed on each block to generate interpolated frames, then frame rate conversion is achieved, but small-object regions cannot be accurately determined leading to quality degradation
Solution Approach 1:
The patent performs preliminary actions by first identifying small-object regions and their motion characteristics before performing smooth filtering. By pre-segmenting the frame and pre-identifying small-object regions, the patent ensures that subsequent processing does not degrade the quality of these regions while still achieving frame rate conversion through interpolated frame generation.
Solution Approach 2:
The patent uses feedback mechanisms to refine small-object region determination. By analyzing motion vector characteristics and comparing them against background motion patterns, the patent iteratively identifies and refines the boundaries of small-object regions, improving measurement precision while maintaining productivity.
Data Source
AI summary
A method and an apparatus for determining a small-object region in a video frame. The method includes dividing a current video frame into at least two regions, and determining a global motion vector corresponding to each region; determining an interframe motion vector of each group of adjacent frames in multiple video frames that include the current video frame and a reference frame of the current video frame; determining a candidate small-object region in the current video frame according to the interframe motion vector of the each group of adjacent frames and the determined global motion vector corresponding to each region; and performing filtering on the candidate small-object region in the current video frame, and determining a region obtained after the filtering as a small-object region in the current video frame.