Research map / Label-efficient and robust learning / Domain adaptation and generalization
SFDATrack: Generalized Source-Free Domain Adaptive Tracking Under Adverse Weather Conditions
Siyuan Yao, Ziqi Wang, Junqi Huang, Ruiqi Yu, Wenqi Ren, 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 DIM module (ablation) | SFDATrack · NAT2021 | AUC: 51.22 -> 53.56 | Improved |
| add HPP module (ablation) | SFDATrack · NAT2021 | AUC: 51.22 -> 53.24 | Improved |
| add both DIM and HPP (ablation) | SFDATrack · NAT2021 | AUC: 51.22 -> 56.78 | Improved |
| add domain-specific alignment (ablation) | SFDATrack · UAVDark70 | AUC: 58.29 -> 59.65 | Improved |
| add domain-invariant alignment (ablation) | SFDATrack · UAVDark70 | AUC: 58.29 -> 59.55 | Improved |
| add both DSA and DIA (ablation) | SFDATrack · UAVDark70 | AUC: 58.29 -> 61.22 | Improved |
| replace Transformer with Uni-Mamba in DIM (ablation) | SFDATrack · UAVDark70 | AUC: 56.77 -> 58.85 | Improved |
| make Mamba bidirectional (Bi-Mamba) (ablation) | SFDATrack · UAVDark70 | AUC: 58.85 -> 60.33 | Improved |
| use full DIM instead of Bi-Mamba (ablation) | SFDATrack · UAVDark70 | AUC: 60.33 -> 61.22 | Improved |
| update EMA each epoch instead of each batch (ablation) | SFDATrack · NAT2021 | AUC: 55.20 -> 56.78 | Improved |
| update EMA each epoch instead of every 5 epochs (ablation) | SFDATrack · NAT2021 | AUC: 54.32 -> 56.78 | Improved |
| balanced DSA/DIA weights 0.5/0.5 instead of 0.1/0.5 (ablation) | SFDATrack · NAT2021 | AUC: 55.09 -> 56.78 | Improved |
Earlier methods this paper compares with
Named baselines from earlier accepted papers in this paper's results tables: better than 1 of 1 on most shared results.
- UMDATrack — UMDATrack: Unified Multi-Domain Adaptive Tracking Under Adverse Weather Conditions (ICCV 2025) · 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
- UAWTrack: Universal 3D Single Object Tracking in Adverse Weather (AAAI 2025 · similarity 0.85)
- LVPTrack: High Performance Domain Adaptive UAV Tracking with Label Aligned Visual Prompt Tuning (AAAI 2025 · similarity 0.85)
- Genuine Knowledge from Practice: Diffusion Test-Time Adaptation for Video Adverse Weather Removal (CVPR 2024 · similarity 0.84)
- End-to-End Video Semantic Segmentation in Adverse Weather using Fusion Blocks and Temporal-Spatial Teacher-Student Learning (NeurIPS 2024 · similarity 0.84)
- Semantic Segmentation in Multiple Adverse Weather Conditions with Domain Knowledge Retention (AAAI 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.
