ADFactory: An Effective Framework for Generalizing Optical Flow With NeRF

Han Ling, Quansen Sun, Yinghui Sun, Xian Xu, Xingfeng Li

Abstract: A significant challenge facing current optical flow meth-ods is the difficulty in generalizing them well to the real world. This is mainly due to the lack of large-scale real-world datasets, and existing self-supervised methods are limited by indirect loss and occlusions, resulting in fuzzy outcomes. To address this challenge, we introduce a novel optical flow training framework: automatic data factory (ADF). ADF only requires RGB images as input to effectively train the optical flow network on the target data do-main. Specifically, we use advanced NeRF technology to reconstruct scenes from photo groups collected by a monoc-ular camera, and then calculate optical flow labels between camera pose pairs based on the rendering results. To elimi-nate erroneous labels caused by defects in the scene reconstructed by NeRF, we screened the generated labels from multiple aspects, such as optical flow matching accuracy, radiation field confidence, and depth consistency. The fil-tered labels can be directly used for network supervision. Experimentally, the generalization ability of ADF on KITTI surpasses existing self-supervised optical flow and monoc-ular scene flow algorithms. In addition, ADF achieves impressive results in real-world zero-point generalization evaluations and surpasses most supervised methods11Code: https://github.com/HanLingsgjk/UnifiedGeneralization.