Pedestrian Tracking Counter Mechanism for Robust ID Switching
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing multi-target pedestrian tracking methods face challenges such as insufficient robustness in tracking identification switching, difficulties in feature matching due to similar re-identification features, and errors in feature matching when pedestrians overlap, leading to issues like missed detections and incorrect tracking.
Innovation Solution
A multi-target pedestrian tracking method that sets a tracking counter for each candidate detection box, updates it based on matching with existing tracking boxes, and converts temporary to confirmed tracking IDs when reaching a threshold, while calculating feature distances and partial intersection-over-unions to improve matching accuracy.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If traditional tracking-by-detection mode is used to achieve multi-target tracking, then the system can operate with simple algorithms, but tracking accuracy deteriorates due to blockings and crossings between pedestrians
Solution Approach 1:
The patent segments the tracking process into multiple independent modules: detection module for candidate boxes, matching module for feature comparison, counter module for tracking state management, and confirmation module for final identification. Each module handles a specific aspect of the tracking task, improving overall accuracy while maintaining manageable complexity through functional decomposition.
Solution Approach 2:
The patent performs preliminary actions by pre-calculating feature distances between candidate boxes and existing tracking boxes before making tracking decisions. The system also pre-establishes matching criteria and counter thresholds, allowing for more accurate and stable tracking identification switching without requiring complex real-time decision-making algorithms.
2Reliability
If feature matching is performed between candidate boxes and tracking boxes, then tracking continuity is improved, but matching errors occur when pedestrians have similar re-identification features
Solution Approach 1:
The patent implements a feedback mechanism through the tracking counter that continuously monitors matching results. When a candidate box matches an existing tracking box, the counter increments; when it doesn't match, the counter decrements. This feedback loop allows the system to accumulate evidence over multiple frames, reducing the impact of single-frame matching errors caused by similar pedestrian features and improving overall tracking continuity and accuracy.
3Speed
If tracking identification is switched based on single-frame detection, then the system responds quickly to new pedestrians, but false identifications occur due to detection errors
Solution Approach 1:
The patent performs preliminary actions by pre-establishing a confirmation threshold for the tracking counter before finalizing identification changes. This threshold mechanism ensures that detection errors in single frames do not trigger false identification switches, while still allowing quick response to genuine new pedestrians once the threshold is reached through cumulative matching evidence.
Solution Approach 2:
The patent applies beforehand cushioning by using the tracking counter to buffer against detection errors. The counter requires multiple consistent matching results before confirming an identification change, providing a cushion against false positives from single-frame detection errors while maintaining responsiveness to actual tracking needs.
4Quantity of substance
If multiple candidate boxes are detected in overlapping regions, then detection coverage is improved, but feature matching becomes ambiguous leading to tracking errors
Solution Approach 1:
The patent uses feedback through the tracking counter to resolve ambiguity in feature matching when multiple candidate boxes overlap. The counter accumulates matching evidence across frames, allowing the system to distinguish between genuine multiple pedestrians and overlapping detection false positives by observing consistent matching patterns over time, thereby improving feature matching precision while maintaining detection coverage.
Data Source
AI summary
A multi-target pedestrian tracking method, a multi-target pedestrian tracking apparatus and a multi-target pedestrian tracking device are provided, related to the field of image processing technologies. The multi-target pedestrian tracking method includes: detecting a plurality of candidate pedestrian detection boxes in a current frame of image to be detected, where a temporary tracking identification and a tracking counter are set for each of the plurality of candidate pedestrian detection boxes; and determining whether each of the plurality of candidate pedestrian detection boxes matches an existing tracking box, updating a value of the tracking counter according to a determination result, and continuing to detect a next frame of image to be detected. When the value of the tracking counter reaches a first preset threshold, the updating the value of the tracking counter is stopped, and the temporary tracking identification is converted to a confirmed tracking identification.