Trajectory Tracking Using Head Image Segmentation
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current motion tracking technologies fail to accurately track human trajectories in videos when a human body is partially blocked, leading to association failures and incomplete motion trajectory analysis.
Innovation Solution
A trajectory tracking system that performs motion tracking on head images within video frames, using head tracking algorithms like model methods, grayscale distribution methods, and block feature methods, to obtain motion trajectories, and combines these with face image sets to determine and synthesize complete motion trajectories for individuals.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If motion tracking is performed on human bodies in video frames, then complete motion trajectories can be obtained, but tracking accuracy deteriorates when human bodies are blocked
Solution Approach 1:
The patent segments the human body detection into two independent parts: head detection and body detection. By focusing on the head region which is less likely to be blocked, the system maintains tracking reliability even when the full body is occluded. The head detection module operates independently to provide robust tracking anchors.
Solution Approach 2:
The patent introduces face image sets as an intermediary element to connect interrupted trajectories. When body detection fails due to blocking, the face images serve as mediators to associate different body detections with the same individual, enabling trajectory continuation without direct body-to-body tracking.
2Productivity
If global target association algorithm is used to associate human bodies in all video frames, then complete motion trajectories can be obtained, but association accuracy deteriorates when human bodies are blocked
Solution Approach 1:
The patent segments the association process into head-based association and body-based association. The head detection provides stable identification across frames, while body detection supplements the trajectory. This segmentation allows the system to maintain association accuracy even when body detection fails due to blocking.
Solution Approach 2:
The patent uses face images as copies or representations of individuals to establish identity consistency. Instead of directly tracking bodies that may be blocked, the system uses face image sets as representative copies to maintain association between different body detections and the same individual.
3Reliability
If motion tracking is performed on head images instead of full bodies, then tracking reliability improves when blocked, but trajectory completeness may deteriorate
Solution Approach 1:
The patent merges head-based tracking and body-based tracking into a unified trajectory system. By combining the reliable head detection with body detection results, the system achieves both reliability (from head tracking) and completeness (from body trajectory data), resolving the trade-off between the two.
Solution Approach 2:
The patent ensures continuity of useful action by using face image sets to bridge interrupted body detections. When body tracking is interrupted due to blocking, the face images provide continuous identification, allowing the trajectory to continue without breaks in the useful tracking action.
Data Source
Figure 1~2a
Figure 2b~2d
Figure 2e~2f
AI summary
Disclosed in embodiments of the present application are a trajectory tracking method and apparatus, and a storage medium. The embodiments of the present application comprise: performing motion tracking on head images in a plurality of video frames to obtain a motion trajectory corresponding to the head images; obtaining face images corresponding to the head images in the different video frames to obtain a face image set corresponding to the head images; determining at least two face image sets having the same face image from the face image set corresponding to the head images; and combining the motion trajectories corresponding to the determined face image sets to obtain a final motion trajectory of the trajectory tracking.