FlowAdapt achieves state-of-the-art domain adaptation for V2X collaborative perception via optimal transport flow, using Wasserstein greedy sampling and progressive knowledge transfer with only 1% trainable parameters.
Opv2v: An open benchmark dataset and fusion pipeline for perception with vehicle-to-vehicle communica- tion, 2022
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.CV 1years
2026 1verdicts
UNVERDICTED 1representative citing papers
citing papers explorer
-
Move What Matters: Parameter-Efficient Domain Adaptation via Optimal Transport Flow for Collaborative Perception
FlowAdapt achieves state-of-the-art domain adaptation for V2X collaborative perception via optimal transport flow, using Wasserstein greedy sampling and progressive knowledge transfer with only 1% trainable parameters.