Learned Dual-View Reflection Removal

Simon Niklaus, Xuaner (Cecilia) Zhang, Jonathan T. Barron, Neal Wadhwa, Rahul Garg, Feng Liu, Tianfan Xue; Proceedings of the IEEE/CVF Winter Conference on Applications of Computer Vision (WACV), 2021, pp. 3713-3722


Traditional reflection removal algorithms either use a single image as input, which suffers from intrinsic ambiguities, or use multiple images from a moving camera, which is inconvenient for users. We instead propose a learning-based dereflection algorithm that uses stereo images as input. This is an effective trade-off between the two extremes: the parallax between two views provides cues to remove reflections, and two views are easy to capture due to the adoption of stereo cameras in smartphones. Our model consists of a learning-based reflection-invariant flow model for dual-view registration, and a learned synthesis model for combining aligned image pairs. Because no dataset for dual-view reflection removal exists, we render a synthetic dataset of dual-views with and without reflections for use in training. Our evaluation on an additional real-world dataset of stereo pairs shows that our algorithm outperforms existing single-image and multi-image dereflection approaches.

Related Material

[pdf] [supp] [arXiv]
@InProceedings{Niklaus_2021_WACV, author = {Niklaus, Simon and Zhang, Xuaner (Cecilia) and Barron, Jonathan T. and Wadhwa, Neal and Garg, Rahul and Liu, Feng and Xue, Tianfan}, title = {Learned Dual-View Reflection Removal}, booktitle = {Proceedings of the IEEE/CVF Winter Conference on Applications of Computer Vision (WACV)}, month = {January}, year = {2021}, pages = {3713-3722} }