Video Key Point Positioning with Historical Region Stabilization
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current methods for object key point positioning in images and videos are time-consuming, labor-intensive, and suffer from low accuracy due to the need for manual detection and the occurrence of jitter between video frames.
Innovation Solution
An object key point positioning method that detects a target object in a current video frame, stabilizes the detection region using a historic detection region, and performs key point positioning to obtain accurate and stable key points, reducing jitter and improving positioning accuracy.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If manual detection is used to position object key points, then the method is simple to implement, but the positioning accuracy is low and the process is time-consuming
Solution Approach 1:
The patent replaces manual mechanical detection with an automated computer vision system that uses deep learning models to detect and position object key points automatically, eliminating the need for manual intervention while achieving high positioning accuracy
Solution Approach 2:
The patent introduces an intermediate processing step that uses detection regions from previous video frames as reference to stabilize and refine key point positions in current frames, improving accuracy without requiring complete re-detection
2Measurement precision
If key point positioning is performed on each video frame independently, then the processing is straightforward, but jitter occurs between frames reducing accuracy
Solution Approach 1:
The patent implements a feedback mechanism where detection results from previous frames are used to guide and stabilize detection in current frames. The system continuously refines key point positions by comparing with historical detection regions, reducing jitter and improving temporal consistency
Solution Approach 2:
The patent performs preliminary detection to obtain detection regions in advance, then uses these pre-obtained regions as references for stabilizing subsequent key point positioning, reducing computational burden and improving stability
Data Source
AI summary
An image processing method and apparatus, and a storage medium are provided. The method includes: detecting a target object in a current video frame of a target video stream, to obtain a current detection region for the target object; adjusting the current detection region according to a historic detection region corresponding to the target object in a historic video frame of the target video stream, to obtain a determined current detection region; performing key point positioning on the target object based on the determined current detection region, to obtain a first set of key points; and performing stabilization on locations of the key points in the first set according to locations of key points in a second set corresponding to the target object in the historic video frame, to obtain current locations of a set of key points of the target object in the current video frame.