PolyTrack: Tracking with Bounding Polygons
This work addresses tracking for automated driving systems, but it is incremental as it adapts existing techniques with polygon-based segmentation.
The paper tackles multi-object tracking and segmentation by introducing PolyTrack, which uses bounding polygons instead of traditional bounding boxes, achieving results that serve as a good alternative to existing methods on MOTS and KITTIMOTS datasets.
In this paper, we present a novel method called PolyTrack for fast multi-object tracking and segmentation using bounding polygons. Polytrack detects objects by producing heatmaps of their center keypoint. For each of them, a rough segmentation is done by computing a bounding polygon over each instance instead of the traditional bounding box. Tracking is done by taking two consecutive frames as input and computing a center offset for each object detected in the first frame to predict its location in the second frame. A Kalman filter is also applied to reduce the number of ID switches. Since our target application is automated driving systems, we apply our method on urban environment videos. We trained and evaluated PolyTrack on the MOTS and KITTIMOTS datasets. Results show that tracking polygons can be a good alternative to bounding box and mask tracking. The code of PolyTrack is available at https://github.com/gafaua/PolyTrack.