Optical flow models trained on high-quality data often degrade severely when confronted with real-world corruptions such as blur, noise, and compression artifacts. To overcome this limitation, we formulate Degradation-Aware Optical Flow, a new task targeting accurate dense correspondence estimation from real-world corrupted videos. Our key insight is that the intermediate representations of image restoration diffusion models are inherently corruption-aware but lack temporal awareness. To address this limitation, we lift the model to attend across adjacent frames via full spatio-temporal attention, and empirically demonstrate that the resulting features exhibit zero-shot correspondence capabilities. Based on this finding, we present DA-Flow, a hybrid architecture that fuses these diffusion features with convolutional features within an iterative refinement framework. DA-Flow substantially outperforms existing optical flow methods under severe degradation across multiple benchmarks.
@article{arxiv.2603.23499,
title = {DA-Flow: Degradation-Aware Optical Flow Estimation with Diffusion Models},
author = {Jaewon Min and Jaeeun Lee and Yeji Choi and Paul Hyunbin Cho and Jin Hyeon Kim and Tae-Young Lee and Jongsik Ahn and Hwayeong Lee and Seonghyun Park and Seungryong Kim},
journal= {arXiv preprint arXiv:2603.23499},
year = {2026}
}