S3R-Net: A Single-Stage Approach to Self-Supervised Shadow Removal

Nikolina Kubiak, Armin Mustafa, Graeme Phillipson, Stephen Jolly, Simon Hadfield; Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR) Workshops, 2024, pp. 5898-5908

In this paper we present S3R-Net the Self-Supervised Shadow Removal Network. The two-branch WGAN model achieves self-supervision relying on the unify-and-adapt phenomenon - it unifies the style of the output data and infers its characteristics from a database of unaligned shadow-free reference images. This approach stands in contrast to the large body of supervised frameworks. S3R-Net also differentiates itself from the few existing self-supervised models operating in a cycle-consistent manner as it is a non-cyclic unidirectional solution. The proposed framework achieves comparable numerical scores to recent self-supervised shadow removal models while exhibiting superior qualitative performance and keeping the computational cost low.