-
[pdf]
[bibtex]@InProceedings{Vasluianu_2023_CVPR, author = {Vasluianu, Florin-Alexandru and Seizinger, Tim and Timofte, Radu and Cui, Shuhao and Huang, Junshi and Tian, Shuman and Fan, Mingyuan and Zhang, Jiaqi and Zhu, Li and Wei, Xiaoming and Wei, Xiaolin and Luo, Ziwei and Gustafsson, Fredrik K. and Zhao, Zheng and Sj\"olund, Jens and Sch\"on, Thomas B. and Dong, Xiaoyi and Zhang, Xi Sheryl and Li, Chenghua and Leng, Cong and Yeo, Woon-Ha and Oh, Wang-Taek and Lee, Yeo-Reum and Ryu, Han-Cheol and Luo, Jinting and Jiang, Chengzhi and Han, Mingyan and Wu, Qi and Lin, Wenjie and Yu, Lei and Li, Xinpeng and Jiang, Ting and Fan, Haoqiang and Liu, Shuaicheng and Xu, Shuning and Song, Binbin and Chen, Xiangyu and Zhang, Shile and Zhou, Jiantao and Zhang, Zhao and Zhao, Suiyi and Zheng, Huan and Gao, Yangcheng and Wei, Yanyan and Wang, Bo and Ren, Jiahuan and Luo, Yan and Kondo, Yuki and Miyata, Riku and Yasue, Fuma and Naruki, Taito and Ukita, Norimichi and Chang, Hua-En and Yang, Hao-Hsiang and Chen, Yi-Chung and Chiang, Yuan-Chun and Huang, Zhi-Kai and Chen, Wei-Ting and Chen, I-Hsiang and Hsieh, Chia-Hsuan and Kuo, Sy-Yen and Xianwei, Li and Fu, Huiyuan and Liu, Chunlin and Ma, Huadong and Fu, Binglan and He, Huiming and Wang, Mengjia and She, Wenxuan and Liu, Yu and Nathan, Sabari and Kansal, Priya and Zhang, Zhongjian and Yang, Huabin and Wang, Yan and Zhang, Yanru and Phutke, Shruti S. and Kulkarni, Ashutosh and Khan, MD Raqib and Murala, Subrahmanyam and Vipparthi, Santosh Kumar and Ye, Heng and Liu, Zixi and Yang, Xingyi and Liu, Songhua and Wu, Yinwei and Jing, Yongcheng and Yu, Qianhao and Zheng, Naishan and Huang, Jie and Long, Yuhang and Yao, Mingde and Zhao, Feng and Zhao, Bowen and Ye, Nan and Shen, Ning and Cao, Yanpeng and Xiong, Tong and Xia, Weiran and Li, Dingwen and Xia, Shuchen}, title = {NTIRE 2023 Image Shadow Removal Challenge Report}, booktitle = {Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR) Workshops}, month = {June}, year = {2023}, pages = {1788-1807} }
NTIRE 2023 Image Shadow Removal Challenge Report
Abstract
This work reviews the results of the NTIRE 2023 Challenge on Image Shadow Removal. The described set of solutions were proposed for a novel dataset, which captures a wide range of object-light interactions. It consists of 1200 roughly pixel aligned pairs of real shadow free and shadow affected images, captured in a controlled environment. The data was captured in a white-box setup, using professional equipment for lights and data acquisition. The challenge had a number of 144 participants registered, out of which 19 teams were compared in the final ranking. The proposed solutions extend the work on shadow removal, improving over well-established state-of-the-art methods.
Related Material