Gr-IoU: Ground-Intersection over Union for Robust Multi-Object Tracking with 3D Geometric Constraints
This addresses tracking errors in crowded or occluded scenes for applications like surveillance or autonomous driving, but it is incremental as it builds on existing IoU methods with geometric enhancements.
The paper tackles the data association problem in multi-object tracking by proposing Ground IoU (Gr-IoU), which uses 3D geometric constraints to transform bounding boxes to the ground plane, improving accuracy and reducing ID switches. It outperforms conventional real-time methods on MOT17 and MOT20 datasets.
We propose a Ground IoU (Gr-IoU) to address the data association problem in multi-object tracking. When tracking objects detected by a camera, it often occurs that the same object is assigned different IDs in consecutive frames, especially when objects are close to each other or overlapping. To address this issue, we introduce Gr-IoU, which takes into account the 3D structure of the scene. Gr-IoU transforms traditional bounding boxes from the image space to the ground plane using the vanishing point geometry. The IoU calculated with these transformed bounding boxes is more sensitive to the front-to-back relationships of objects, thereby improving data association accuracy and reducing ID switches. We evaluated our Gr-IoU method on the MOT17 and MOT20 datasets, which contain diverse tracking scenarios including crowded scenes and sequences with frequent occlusions. Experimental results demonstrated that Gr-IoU outperforms conventional real-time methods without appearance features.