Research map / Label-efficient and robust learning / Domain adaptation and generalization
UMDATrack: Unified Multi-Domain Adaptive Tracking Under Adverse Weather Conditions
Siyuan Yao, Rui Zhu, Ziqi Wang, Wenqi Ren, Yanyang Yan, Xiaochun Cao
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 controllable scenario generator (CSG) data (ablation) | baseline tracker · NAT2021 (ablation) | AUC: 49.11 -> 50.90 | Improved |
| add domain-customized adapter (DCA) (ablation) | baseline tracker · NAT2021 (ablation) | AUC: 49.11 -> 50.56 | Improved |
| add TCA to CSG (ablation) | UMDATrack · NAT2021 | AUC: 50.90 -> 52.27 | Improved |
| add DCA to CSG+TCA (ablation) | UMDATrack · NAT2021 | AUC: 52.27 -> 54.58 | Improved |
| add TCA to CSG+DCA (ablation) | UMDATrack · NAT2021 | precision: 67.49 -> 70.78 | Improved |
| update EMA each epoch instead of every 5 epochs (ablation) | UMDATrack · NAT2021 | AUC: 52.90 -> 54.48 | Improved |
| update EMA each epoch instead of each batch (ablation) | UMDATrack · NAT2021 | AUC: 53.89 -> 54.48 | Improved |
| raise synthetic dataset sampling ratio from 1 to 4 (ablation) | UMDATrack · NAT2021 | AUC: 53.13 -> 54.58 | Improved |
| raise synthetic dataset sampling ratio from 4 to 6 (ablation) | UMDATrack · NAT2021 | AUC: 54.58 -> 54.26 | No clear effect |
| add text prompts to CSG (ablation) | CSG generator · NAT2021 | AUC: 52.53 -> 54.58 | Improved |
| use CSG instead of CycleGAN (ablation) | data generator · NAT2021 | AUC: 51.10 -> 54.58 | Improved |
| use CSG instead of UNIT (ablation) | data generator · NAT2021 | AUC: 50.23 -> 54.58 | Improved |
Later papers that compared with this method
1 later accepted paper reports UMDATrack as a baseline in their results tables; 1 of them report a better number on most of the results they share with it.
- SFDATrack: Generalized Source-Free Domain Adaptive Tracking Under Adverse Weather Conditions (ECCV 2026 · better on 11 of 12 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
- LVPTrack: High Performance Domain Adaptive UAV Tracking with Label Aligned Visual Prompt Tuning (AAAI 2025 · similarity 0.86)
- UAWTrack: Universal 3D Single Object Tracking in Adverse Weather (AAAI 2025 · similarity 0.86)
- UCMCTrack: Multi-Object Tracking with Uniform Camera Motion Compensation (AAAI 2024 · similarity 0.83)
- End-to-End Video Semantic Segmentation in Adverse Weather using Fusion Blocks and Temporal-Spatial Teacher-Student Learning (NeurIPS 2024 · similarity 0.83)
- Genuine Knowledge from Practice: Diffusion Test-Time Adaptation for Video Adverse Weather Removal (CVPR 2024 · similarity 0.83)
- Parsing All Adverse Scenes: Severity-Aware Semantic Segmentation with Mask-Enhanced Cross-Domain Consistency (AAAI 2024 · similarity 0.83)
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.
