2019/03/31 by Xiaowei Hu, Yitong Jiang, Chi-Wing Fu +1 · 216 citations
Computer Science · #Advanced Image Processing Techniques #Adversarial system #Generative Adversarial Networks and Image Synthesis #Human Pose and Action Recognition #Image (mathematics) #Key (lock) #Shadow (psychology) #Shadow mask #cs.CV #cs.MM
paper · pdf · doi:10.1109/iccv.2019.00256
published as IEEE International Conference on Computer Vision (ICCV), pp. 2472-2481, 2019 · Accepted to ICCV 2019
openalex created_date 2019/04/01 · arxiv created 2019/08/08 · openalex publication_date 2019/10/01 · arxiv updated 2020/05/19 · openalex updated_date 2026/08/05
This paper presents a new method for shadow removal using unpaired data, enabling us to avoid tedious annotations and obtain more diverse training samples. However, directly employing adversarial learning and cycle-consistency constraints is insufficient to learn the underlying relationship between the shadow and shadow-free domains, since the mapping between shadow and shadow-free images is not simply one-to-one. To address the problem, we formulate Mask-ShadowGAN, a new deep framework that automatically learns to produce a shadow mask from the input shadow image and then takes the mask to guide the shadow generation via re-formulated cycle-consistency constraints. Particularly, the framework simultaneously learns to produce shadow masks and learns to remove shadows, to maximize the overall performance. Also, we prepared an unpaired dataset for shadow removal and demonstrated the effectiveness of Mask-ShadowGAN on various experiments, even it was trained on unpaired data.