UnShadowNet: Illumination Critic Guided Contrastive Learning for Shadow Removal
Article Type
Research Article
Publication Title
IEEE Access
Abstract
Shadows are frequently encountered natural phenomena that significantly hinder the performance of computer vision perception systems in practical settings, e.g., autonomous driving. A solution to this would be to eliminate shadow regions from the images before the processing of the perception system. Yet, training such a solution requires pairs of aligned shadowed and non-shadowed images which are difficult to obtain. We introduce a novel weakly supervised shadow removal framework UnShadowNet trained using contrastive learning. It is composed of a DeShadower network responsible for the removal of the extracted shadow under the guidance of an Illumination network which is trained adversarially by the illumination critic and a Refinement network to further remove artefacts. We show that UnShadowNet can be easily extended to a fully-supervised set-up to exploit the ground-truth when available. UnShadowNet outperforms existing state-of-the-art approaches on three publicly available shadow datasets (ISTD, adjusted ISTD, SRD) in both the weakly and fully supervised setups.
First Page
87760
Last Page
87774
DOI
https://10.1109/ACCESS.2023.3305576
Publication Date
1-1-2023
Recommended Citation
Dasgupta, Subhrajyoti; Das, Arindam; Yogamani, Senthil; Das, Sudip; Eising, Ciaran; Bursuc, Andrei; and Bhattacharya, Ujjwal, "UnShadowNet: Illumination Critic Guided Contrastive Learning for Shadow Removal" (2023). Journal Articles. 3936.
https://digitalcommons.isical.ac.in/journal-articles/3936
Comments
Open Access, Gold