Research map / Detection and segmentation / Tracking
Tracking without Label: Unsupervised Multiple Object Tracking via Contrastive Similarity Learning
Sha Meng, Dian Shao, Jiacheng Guo, Shan Gao
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 |
|---|---|---|---|
| add direct and indirect self-contrast loss over CycAs baseline (ablation) | UCSL · MOT17 test | IDF1: 59.1 -> 67.2 | Improved |
| add cross-contrast loss on top of self-contrast (ablation) | UCSL · MOT17 test | IDF1: 67.2 -> 68.4 | Improved |
| add ambiguity contrast loss (ablation) | UCSL · MOT17 test | MOTA: 69.8 -> 70.5 | Improved |
| add ambiguity contrast loss (ablation) | UCSL · MOT17 test | IDF1: 68.4 -> 68.2 | No clear effect |
| add ambiguity contrast loss (ablation) | UCSL · MOT17 test | FP: 32619 -> 40569 | Hurt |
| use consecutive frames (interval 1) instead of interval 7 (ablation) | UCSL · MOT17 test | MOTA: 68.5 -> 70.5 | Improved |
| increase ReID dimension from 128 to 256 (ablation) | UCSL · MOT17 test | MOTA: 70.5 -> 70.7 | No clear effect |
| apply UCSL to JDE yolov5s instead of supervised ReID | JDE · MOT17 test | HOTA: HOTA 54.1 -> 55.7, MOTA 70.2 -> 69.6 | Mixed |
| use unsupervised UCSL instead of supervised ReID training in FairMOT | FairMOT · MOT17 test | IDF1: 72.3 -> 70.4 | Hurt |
| use UCSL instead of UTrack ReID learning | FairMOT+CycAs · MOT17 test | MOTA: 71.8 -> 73.0 | Improved |
Closest papers by meaning
- From Detection to Association: Learning Discriminative Object Embeddings for Multi-Object Tracking (CVPR 2026 · similarity 0.89)
- Self-Supervised Multi-Object Tracking with Cross-input Consistency (NeurIPS 2021 · similarity 0.88)
- QDTrack: Quasi-Dense Similarity Learning for Appearance-Only Multiple Object Tracking (TPAMI 2023 · similarity 0.88)
- Video Diffusion Models Excel at Tracking Similar-Looking Objects Without Supervision (NeurIPS 2025 · similarity 0.87)
- DiffusionTrack: Diffusion Model for Multi-Object Tracking (AAAI 2024 · similarity 0.87)
- Self-Supervised Multi-Object Tracking with Path Consistency (CVPR 2024 · similarity 0.87)
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.
