Research map / Detection and segmentation / Tracking
SOVTrack: Open-Vocabulary Multi-Object Tracking with Self-Supervised Pseudo Labeling and Feature Distillation
Zekun QIAN, Ruize Han, Junhui Hou, Wei Feng
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 cycle consistency filtering (ablation) | SOVTrack · TAO val | novel TETA: 32.2 -> 35.7 | Improved |
| add cycle consistency filtering (ablation) | SOVTrack · TAO val | base TETA: 38.1 -> 40.1 | Improved |
| use detector prompts instead of SAM auto-segmentation prompts (ablation) | SOVTrack · TAO val | novel TETA: 32.5 -> 35.7 | Improved |
| add adversarial branch (ablation) | SOVTrack · TAO val | novel TETA: 33.0 -> 35.7 | Improved |
| add contrastive branch (ablation) | SOVTrack · TAO val | base TETA: 38.8 -> 40.1 | Improved |
| add adaptive feature fusion (ablation) | SOVTrack · TAO val | novel TETA: 33.2 -> 35.7 | Improved |
| add localization cue (ablation) | SOVTrack · TAO val | novel TETA: 34.5 -> 35.7 | Improved |
| add association cue (ablation) | SOVTrack · TAO val | novel TETA: 34.2 -> 35.7 | Improved |
| add classification cue (ablation) | SOVTrack · TAO val | novel TETA: 33.7 -> 35.7 | Improved |
| add adversarial distillation (vs plain L2 distillation) (ablation) | SOVTrack · TAO val | novel TETA: 34.2 -> 35.7 | Improved |
| add fusion loss (ablation) | SOVTrack · TAO val | novel TETA: 34.6 -> 35.7 | Improved |
| add localization loss from SAM2 masks (ablation) | SOVTrack · TAO val | novel TETA: 32.8 -> 35.7 | Improved |
Earlier methods this paper compares with
Named baselines from earlier accepted papers in this paper's results tables: better than 4 of 4 on most shared results.
- OVTR — OVTR: End-to-End Open-Vocabulary Multiple Object Tracking with Transformer (ICLR 2025) · better on 4 of 4 shared results
- OVTrack — OVTrack: Open-Vocabulary Multiple Object Tracking (CVPR 2023) · better on 4 of 4 shared results
- QDTrack — QDTrack: Quasi-Dense Similarity Learning for Appearance-Only Multiple Object Tracking (TPAMI 2023) · better on 4 of 4 shared results
- ByteTrack — ByteTrack: Multi-Object Tracking by Associating Every Detection Box (ECCV 2022) · better on 2 of 2 shared results
Matched by method name in results tables read by AI. Authors' own tables tend to favour the newer method, and a comparison can be missed when a method is renamed.
Closest papers by meaning
- COVTrack++: Learning Open-Vocabulary Multi-Object Tracking from Continuous Videos via a Synergistic Paradigm (TPAMI 2026 · similarity 0.93)
- OVTrack: Open-Vocabulary Multiple Object Tracking (CVPR 2023 · similarity 0.92)
- OVTR: End-to-End Open-Vocabulary Multiple Object Tracking with Transformer (ICLR 2025 · similarity 0.90)
- VastTrack: Vast Category Visual Object Tracking (NeurIPS 2024 · similarity 0.90)
- OVT-B: A New Large-Scale Benchmark for Open-Vocabulary Multi-Object Tracking (NeurIPS 2024 · similarity 0.89)
- SUTrack: Towards Simple and Unified Single Object Tracking (AAAI 2025 · similarity 0.89)
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.
