Object Tracking via Heat Map Density Analysis
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing object tracking technologies face challenges when multiple objects are close to each other, leading to dense tracking results where multiple objects are recognized as one large object.
Innovation Solution
An object tracking method and device that extracts historical moving traces and predicts object locations, using a heat map to determine object states and comparing similarity values between predicted and current object boxes to accurately track objects.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If object tracking is performed using conventional methods, then tracking coverage is provided, but multiple objects close to each other are recognized as one large object leading to dense tracking results
Solution Approach 1:
The patent applies segmentation by dividing the tracking process into multiple stages: initial object detection, heat map generation to identify dense regions, and secondary detection within those regions. This multi-level segmentation allows the system to distinguish individual objects within densely packed areas that would otherwise be merged into a single detection, thereby improving measurement precision while accurately capturing the true quantity of objects
Solution Approach 2:
The patent introduces a heat map dimension that visualizes object density across the image space. By adding this thermal dimension overlay to the standard object detection process, the system can identify regions where multiple objects are closely packed and apply enhanced detection algorithms specifically to those areas, enabling accurate separation and counting of individual objects without affecting overall tracking coverage
2Reliability
If object boxes are densely packed in static states, then comprehensive coverage is achieved, but misjudgment occurs where multiple objects are confused
Solution Approach 1:
The patent introduces the heat map as an intermediary element that mediates between raw object detections and final tracking decisions. The heat map processes and visualizes spatial density information, allowing the system to identify when multiple objects are densely packed in static regions. This intermediary representation enables the tracker to apply appropriate disambiguation strategies, such as adjusting detection thresholds or applying motion-based separation, thereby maintaining reliable tracking while improving object identification accuracy in challenging dense static scenarios
Data Source
AI summary
An object tracking method includes: extracting historical moving traces corresponding to historical objects from historical images, predicting predicted locations and predicted object boxes of the historical objects, determining the historical objects are in a static state or a moving state according to a heat map, wherein the heat map is generated according to the historical images, extracting current bounding boxes corresponding to current objects from current images, comparing and calculating similarity values between the predicted object boxes and the current bounding boxes respectively, corresponding one of the current objects to one of the historical objects when the similarity value is higher than a threshold value, generating a labelled object box, and using the labelled object box to update the heat map and at least one of the historical moving traces, wherein the labelled object box is in the static state or the moving state.