Research map / Detection and segmentation / Tracking
Assignment-Space-Based Multi-Object Tracking and Segmentation
Anwesa Choudhuri, Girish Chowdhary, Alexander G. Schwing
Read the paperCheck your idea against 137,217 papers
AI-extracted. The measured effects below were read by AI from the paper's full text (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.
Measured effects reported in the paper
| What was changed | Where | Measured | Effect |
|---|---|---|---|
| replace local pairwise assignment with dynamic programming over assignment space (ablation) | Ours (WL) · KITTI-MOTS val (car), PointTrack detections | HOTA: 70.1 -> 81.3 | Improved |
| add long-range tracklet assignment post-processing (ablation) | Ours · KITTI-MOTS val (car), PointTrack detections | HOTA: 81.3 -> 83.4 | Improved |
| add joint training of detector and tracking parameters (ablation) | Ours · KITTI-MOTS val (car), PointTrack detections | HOTA: 83.4 -> 85.0 | Improved |
| add joint training of detector and tracking parameters (ablation) | Ours · KITTI-MOTS val (car), PointTrack detections | AssA: 81.8 -> 86.1 | Improved |
| increase top-k assignments from 1 to 2 (ablation) | Ours · KITTI-MOTS val (car), PointTrack detections | AssA: 76.4 -> 79.9 | Improved |
| increase k from 2 to 15 (ablation) | Ours · KITTI-MOTS val (car), PointTrack detections | HOTA: 82.5 -> 83.4 | Improved |
| increase k from 15 to 20 (ablation) | Ours · KITTI-MOTS val (car), PointTrack detections | HOTA: 83.4 -> 83.4 | No clear effect |
| add IoU term to the assignment cost (ablation) | Ours (WL) · KITTI-MOTS val (car), PointTrack detections | HOTA: 80.5 -> 81.3 | Improved |
| add appearance term to the assignment cost (ablation) | Ours (WL) · KITTI-MOTS val (car), PointTrack detections | HOTA: 80.6 -> 81.3 | Improved |
| add distance term to the assignment cost (ablation) | Ours (WL) · KITTI-MOTS val (car), PointTrack detections | ID switches: 100 -> 77 | Improved |
| add distance term to the assignment cost (ablation) | Ours (WL) · KITTI-MOTS val (car), PointTrack detections | HOTA: 81.2 -> 81.3 | No clear effect |
| use assignment-space tracking vs PointTrack | Ours (JT) · KITTI-MOTS test (car) | AssA: 48.83 -> 62.24 | Improved |
Closest papers by meaning
- MOTS: Multi-Object Tracking and Segmentation (CVPR 2019 · similarity 0.91)
- Multi-Object Tracking and Segmentation Via Neural Message Passing (IJCV 2022 · similarity 0.89)
- Segment as Points for Efficient Online Multi-Object Tracking and Segmentation (ECCV 2020 · similarity 0.89)
- SAM2MOT: A Novel Paradigm of Multi-Object Tracking by Segmentation (AAAI 2026 · similarity 0.88)
- Multiple Object Tracking as ID Prediction (CVPR 2025 · similarity 0.88)
- ByteTrack: Multi-Object Tracking by Associating Every Detection Box (ECCV 2022 · similarity 0.88)
Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.
