UMDATrack unifies multi-weather domain adaptation for visual tracking, using a diffusion-based scenario generator, a domain adapter, and an optimal-transport confidence alignment loss to reach reported state-of-the-art results on night, fog, and rain benchmarks.
Learning discriminative model prediction for track- ing
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.CV 1years
2025 1verdicts
CONDITIONAL 1representative citing papers
citing papers explorer
-
UMDATrack: Unified Multi-Domain Adaptive Tracking Under Adverse Weather Conditions
UMDATrack unifies multi-weather domain adaptation for visual tracking, using a diffusion-based scenario generator, a domain adapter, and an optimal-transport confidence alignment loss to reach reported state-of-the-art results on night, fog, and rain benchmarks.