Research map / Image generation / Text-to-image diffusion
Leveraging Text-to-Image Diffusion Models for Unsupervised Visual Object Tracking
Zhengbo Zhang, Zhigang Tu, Junsong Yuan, De Wen Soh, Bo Du
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 (ablation) | Ours · VOT2018 | EAO: 37.1 -> 38.4 | Improved |
| add online prompt updater (ablation) | Ours · VOT2018 | EAO: 35.5 -> 38.4 | Improved |
| add target-shared embedding E_sh (ablation) | Ours · LaSOT | Success: 49.1 -> 50.4 | Improved |
| add attention map fusion head over multiple UNet layers (ablation) | Ours · LaSOT | Success: 49.5 -> 50.4 | Improved |
| add frequency-domain motion information (ablation) | Ours · LaSOT | Precision: 48.3 -> 49.2 | Improved |
| add long-term motion information (ablation) | Ours · LaSOT | Success: 47.1 -> 50.4 | Improved |
| add short-term motion information (ablation) | Ours · LaSOT | Success: 48.4 -> 50.4 | Improved |
| replace Stable Diffusion V2.1 with PixArt backbone | Ours · LaSOT | Success: 50.4 -> 50.5 | No clear effect |
| replace Stable Diffusion V2.1 with FLUX backbone | Ours · LaSOT | Success: 50.4 -> 50.7 | Mixed |
| diffusion-based prompt tracking versus prior best unsupervised ULAST | Ours · TrackingNet | Success: 65.4 -> 70.4 | Improved |
| diffusion-based prompt tracking versus Diff-Tracker | Ours · LaSOT | Success: 48.6 -> 50.4 | Improved |
Closest papers by meaning
- Emergent Temporal Correspondences from Video Diffusion Transformers (NeurIPS 2025 · similarity 0.90)
- Video Diffusion Models Excel at Tracking Similar-Looking Objects Without Supervision (NeurIPS 2025 · similarity 0.90)
- DiffusionTrack: Diffusion Model for Multi-Object Tracking (AAAI 2024 · similarity 0.89)
- DINTR: Tracking via Diffusion-based Interpolation (NeurIPS 2024 · similarity 0.88)
- Prompt-Free Diffusion: Taking "Text" out of Text-to-Image Diffusion Models (CVPR 2024 · similarity 0.88)
- VidProM: A Million-scale Real Prompt-Gallery Dataset for Text-to-Video Diffusion Models (NeurIPS 2024 · 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.
