Research map / Image generation / Text-to-image diffusion
Diff-Tracker: Text-to-Image Diffusion Models are Unsupervised Trackers
Zhengbo Zhang, Li Xu, Duo Peng, Hossein Rahmani, Jun Liu
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 attention harmonization of self- and cross-attention maps (ablation) | Diff-Tracker · VOT2018 | EAO: 0.359 -> 0.365 | Improved |
| add online prompt updater (ablation) | Diff-Tracker · VOT2018 | EAO: 0.349 -> 0.365 | Improved |
| add online prompt updater (robustness) (ablation) | Diff-Tracker · VOT2018 | Rob (lower is better): 0.292 -> 0.273 | Improved |
| add long-term motion information to prompt updater (ablation) | Diff-Tracker · VOT2018 | EAO: 0.355 -> 0.365 | Improved |
| add short-term motion information to prompt updater (ablation) | Diff-Tracker · VOT2018 | EAO: 0.360 -> 0.365 | Improved |
| use Diff-Tracker vs best prior unsupervised tracker ULAST*-on | unsupervised tracking · TrackingNet | success: 0.654 -> 0.675 | Improved |
| use Diff-Tracker vs ULAST*-on on LaSOT | unsupervised tracking · LaSOT | success: 0.471 -> 0.486 | Improved |
| compare unsupervised Diff-Tracker with supervised trackers | Diff-Tracker · VOT2018 | EAO: 0.365 vs 0.440 (DiMP) | Hurt |
Closest papers by meaning
- Video Diffusion Models Excel at Tracking Similar-Looking Objects Without Supervision (NeurIPS 2025 · similarity 0.90)
- Emergent Temporal Correspondences from Video Diffusion Transformers (NeurIPS 2025 · similarity 0.90)
- DiffusionTrack: Diffusion Model for Multi-Object Tracking (AAAI 2024 · similarity 0.88)
- DiffusionTrack: Point Set Diffusion Model for Visual Object Tracking (CVPR 2024 · similarity 0.88)
- DINTR: Tracking via Diffusion-based Interpolation (NeurIPS 2024 · similarity 0.87)
- Diff-Plugin: Revitalizing Details for Diffusion-based Low-level Tasks (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.
