Video Object Tracking Through Packet-Loss Trajectory Prediction
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing object tracking systems fail to accurately track objects in videos received via networks due to packet loss, leading to missing frames or regions, which disrupt the tracking process.
Innovation Solution
A video processing system that includes a detection unit to identify tracking targets, a trajectory prediction unit to forecast the target's path, and a tracking unit to continue tracking in missing regions using prediction results.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Adaptability or versatility
If video is transmitted via network, then video distribution capability is improved, but packet loss causes missing frames and disrupts tracking
Solution Approach 1:
The system performs preliminary actions by detecting objects in valid frames before packet loss occurs, stores their trajectory information, and predicts future positions. When packet loss is detected, the pre-computed trajectory predictions are immediately applied to maintain continuous tracking without interruption, thus resolving the contradiction between network distribution and tracking reliability.
Solution Approach 2:
The system creates copies of trajectory information from valid frames and uses these copied predictions to fill in missing data during packet loss. Instead of relying on the original lost video frames, the system generates duplicate trajectory estimates based on historical motion patterns, ensuring tracking continuity despite network transmission failures.
2Device complexity
If traditional tracking is used, then tracking is simple, but tracking fails when frames are missing
Solution Approach 1:
The system introduces trajectory prediction as an intermediary mechanism between object detection and tracking. When packet loss occurs, this intermediary step generates predicted position information that bridges the gap caused by missing frames, allowing the tracking system to maintain continuity without requiring complex re-tracking algorithms or additional sensors.
Solution Approach 2:
The system dynamically changes tracking parameters based on video quality conditions. When packet loss is detected, it switches from standard frame-by-frame tracking to prediction-based tracking using stored trajectory parameters. This parameter adaptation allows the system to maintain reliable tracking under degraded network conditions while keeping the base system relatively simple.
3Measurement precision
If prediction is used to fill missing regions, then tracking accuracy is improved, but processing complexity increases
Solution Approach 1:
The system applies partial prediction actions only to the specific regions and frames where packet loss occurs, rather than processing the entire video stream with complex prediction algorithms. By limiting prediction to only the necessary missing regions and using simple linear interpolation based on recent trajectory data, it achieves improved tracking accuracy without proportionally increasing overall processing complexity.
Data Source
AI summary
Provided is a video processing system including: a detection means for detecting a tracking target from an input video; a trajectory prediction means for predicting a trajectory of the tracking target in the video; and a tracking means for tracking the tracking target, and when there is a missing region in the video, for estimating a position of the tracking target in the missing region using a prediction result of the trajectory prediction means. The video processing system may further include a determination means for determining whether or not the tracking target tracked by the tracking means is the same as the tracking target detected by the detection means; and a trajectory information output means for assigning identification information to the detected tracking target.