NTIRE 2025 Challenge on Single Image Reflection Removal in the Wild: Datasets, Methods and Results

Kangning Yang; Jie Cai; Ling Ouyang; Florin Alexandru Vasluianu; Radu Timofte; Jiaming Ding; Huiming Sun; Lan Fu; Jinlong Li; Chiu Man Ho; Zibo Meng; Mingjia Li; Hainuo Wang; Qiming Hu; Jiarui Wang; Hao Zhao; Jin Hu; Xiaojie Guo; Mengru Yang; Kui Jiang; Jin Guo; Yiang Chen; Junjun Jiang; Jing He; Yiqing Wang; Kexin Zhang; Licheng Jiao; Lingling Li; Fang Liu; Wenping Ma; Zhiyang Chen; Hao Fang; Wei Zhang; Runmin Cong; Dheeraj Damodhar Hegde; Jatin Kalal; Nikhil Akalwadi; Ramesh Ashok Tabib; Uma Mudenagudi; Yu Fan Lin; Chia Ming Lee; Chih Chung Hsu; Mengxin Zhang; Xiaochao Qu; Luoqi Liu; Ting Liu; Jinshan Chen; Shu Yu He; Sabari Nathan; K. Uma; A. Sasithradevi; B. Sathya Bama; S. Mohamed Mansoor Roomi; Bilel Benjdira; Anas M. Ali; Wadii Boulila; Wei Dong; Yunzhe Li; Ali Hussein; Han Zhou; Jun Chen; Zeyu Xiao; Zhuoyuan Li

doi:10.1109/CVPRW67362.2025.00122

NTIRE 2025 Challenge on Single Image Reflection Removal in the Wild: Datasets, Methods and Results

Kangning Yang^*
, Jie Cai^*
, Ling Ouyang^*
, Florin Alexandru Vasluianu^*
, Radu Timofte^*
, Jiaming Ding^*
, Huiming Sun^*
, Lan Fu^*
, Jinlong Li^*
, Chiu Man Ho^*
, Zibo Meng^*
, Mingjia Li
, Hainuo Wang
, Qiming Hu
, Jiarui Wang
, Hao Zhao
, Jin Hu
, Xiaojie Guo
, Mengru Yang
, Kui Jiang

Jin Guo, Yiang Chen, Junjun Jiang, Jing He, Yiqing Wang, Kexin Zhang, Licheng Jiao, Lingling Li, Fang Liu, Wenping Ma, Zhiyang Chen, Hao Fang, Wei Zhang, Runmin Cong, Dheeraj Damodhar Hegde, Jatin Kalal, Nikhil Akalwadi, Ramesh Ashok Tabib, Uma Mudenagudi, Yu Fan Lin, Chia Ming Lee, Chih Chung Hsu, Mengxin Zhang, Xiaochao Qu, Luoqi Liu, Ting Liu, Jinshan Chen, Shu Yu He, Sabari Nathan, K. Uma, A. Sasithradevi, B. Sathya Bama, S. Mohamed Mansoor Roomi, Bilel Benjdira, Anas M. Ali, Wadii Boulila, Wei Dong, Yunzhe Li, Ali Hussein, Han Zhou, Jun Chen, Zeyu Xiao, Zhuoyuan Li

^*Corresponding author for this work

Faculty of Computing

OPPO AI Center
University of Würzburg
Tianjin University
Harbin Institute of Technology
Xidian University
Shandong University
KLE Technological University
National Cheng Kung University
Beijing meitu company
Couger Inc
Jawaharlal Nehru Technological University Hyderabad
Vellore Institute of Technology
Thiagarajar College of Engineering
Prince Sultan University (PSU)
McMaster University
National University of Singapore
University of Science and Technology of China

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution › peer-review

30 Scopus citations

Abstract

In this paper, we review the NTIRE 2025 challenge on single-image reflection removal (SIRR) in the wild. SIRR is a fundamental task in image restoration. Despite progress in academic research, most methods are tested on synthetic images or limited real-world images, creating a gap in realworld applications. In this challenge, participants are required to process real-world images that cover a range of reflection scenarios and intensities, with the goal of generating clean images without reflections. The challenge attracted more than 200 registrations, with 11 of them participating in the final testing phase. The top-ranked methods advanced the state-of-the-art reflection removal performance and earned unanimous recognition from the five experts in the field. The proposed datasets are available at https://huggingface.co/datasets/qiuzhangTiTi/NTIRE2025-SIRR and the homepage of this challenge is at https://github.com/caijie0620/Reflection-Removal-in-thewild.

Original language	English
Title of host publication	Proceedings - 2025 IEEE/CVF Conference on Computer Vision and Pattern Recognition Workshops, CVPRW 2025
Publisher	IEEE Computer Society
Pages	1292-1302
Number of pages	11
ISBN (Electronic)	9798331599942
DOIs	https://doi.org/10.1109/CVPRW67362.2025.00122
State	Published - 2025
Event	2025 IEEE/CVF Conference on Computer Vision and Pattern Recognition Workshops, CVPRW 2025 - Nashville, United States Duration: 11 Jun 2025 → 12 Jun 2025

Publication series

Name	IEEE Computer Society Conference on Computer Vision and Pattern Recognition Workshops
ISSN (Print)	2160-7508
ISSN (Electronic)	2160-7516

Conference

Conference	2025 IEEE/CVF Conference on Computer Vision and Pattern Recognition Workshops, CVPRW 2025
Country/Territory	United States
City	Nashville
Period	11/06/25 → 12/06/25

Access to Document

10.1109/CVPRW67362.2025.00122

Cite this

Yang, K., Cai, J., Ouyang, L., Vasluianu, F. A., Timofte, R., Ding, J., Sun, H., Fu, L., Li, J., Ho, C. M., Meng, Z., Li, M., Wang, H., Hu, Q., Wang, J., Zhao, H., Hu, J., Guo, X., Yang, M., ... Li, Z. (2025). NTIRE 2025 Challenge on Single Image Reflection Removal in the Wild: Datasets, Methods and Results. In Proceedings - 2025 IEEE/CVF Conference on Computer Vision and Pattern Recognition Workshops, CVPRW 2025 (pp. 1292-1302). (IEEE Computer Society Conference on Computer Vision and Pattern Recognition Workshops). IEEE Computer Society. https://doi.org/10.1109/CVPRW67362.2025.00122

Yang, Kangning ; Cai, Jie ; Ouyang, Ling et al. / NTIRE 2025 Challenge on Single Image Reflection Removal in the Wild : Datasets, Methods and Results. Proceedings - 2025 IEEE/CVF Conference on Computer Vision and Pattern Recognition Workshops, CVPRW 2025. IEEE Computer Society, 2025. pp. 1292-1302 (IEEE Computer Society Conference on Computer Vision and Pattern Recognition Workshops).

@inproceedings{c48096f87ed24fff999e3c40893d84b8,

title = "NTIRE 2025 Challenge on Single Image Reflection Removal in the Wild: Datasets, Methods and Results",

abstract = "In this paper, we review the NTIRE 2025 challenge on single-image reflection removal (SIRR) in the wild. SIRR is a fundamental task in image restoration. Despite progress in academic research, most methods are tested on synthetic images or limited real-world images, creating a gap in realworld applications. In this challenge, participants are required to process real-world images that cover a range of reflection scenarios and intensities, with the goal of generating clean images without reflections. The challenge attracted more than 200 registrations, with 11 of them participating in the final testing phase. The top-ranked methods advanced the state-of-the-art reflection removal performance and earned unanimous recognition from the five experts in the field. The proposed datasets are available at https://huggingface.co/datasets/qiuzhangTiTi/NTIRE2025-SIRR and the homepage of this challenge is at https://github.com/caijie0620/Reflection-Removal-in-thewild.",

author = "Kangning Yang and Jie Cai and Ling Ouyang and Vasluianu, \{Florin Alexandru\} and Radu Timofte and Jiaming Ding and Huiming Sun and Lan Fu and Jinlong Li and Ho, \{Chiu Man\} and Zibo Meng and Mingjia Li and Hainuo Wang and Qiming Hu and Jiarui Wang and Hao Zhao and Jin Hu and Xiaojie Guo and Mengru Yang and Kui Jiang and Jin Guo and Yiang Chen and Junjun Jiang and Jing He and Yiqing Wang and Kexin Zhang and Licheng Jiao and Lingling Li and Fang Liu and Wenping Ma and Zhiyang Chen and Hao Fang and Wei Zhang and Runmin Cong and Hegde, \{Dheeraj Damodhar\} and Jatin Kalal and Nikhil Akalwadi and Tabib, \{Ramesh Ashok\} and Uma Mudenagudi and Lin, \{Yu Fan\} and Lee, \{Chia Ming\} and Hsu, \{Chih Chung\} and Mengxin Zhang and Xiaochao Qu and Luoqi Liu and Ting Liu and Jinshan Chen and He, \{Shu Yu\} and Sabari Nathan and K. Uma and A. Sasithradevi and Bama, \{B. Sathya\} and \{Mansoor Roomi\}, \{S. Mohamed\} and Bilel Benjdira and Ali, \{Anas M.\} and Wadii Boulila and Wei Dong and Yunzhe Li and Ali Hussein and Han Zhou and Jun Chen and Zeyu Xiao and Zhuoyuan Li",

note = "Publisher Copyright: {\textcopyright} 2025 IEEE.; 2025 IEEE/CVF Conference on Computer Vision and Pattern Recognition Workshops, CVPRW 2025 ; Conference date: 11-06-2025 Through 12-06-2025",

year = "2025",

doi = "10.1109/CVPRW67362.2025.00122",

language = "英语",

series = "IEEE Computer Society Conference on Computer Vision and Pattern Recognition Workshops",

publisher = "IEEE Computer Society",

pages = "1292--1302",

booktitle = "Proceedings - 2025 IEEE/CVF Conference on Computer Vision and Pattern Recognition Workshops, CVPRW 2025",

address = "美国",

}

Yang, K, Cai, J, Ouyang, L, Vasluianu, FA, Timofte, R, Ding, J, Sun, H, Fu, L, Li, J, Ho, CM, Meng, Z, Li, M, Wang, H, Hu, Q, Wang, J, Zhao, H, Hu, J, Guo, X, Yang, M, Jiang, K, Guo, J, Chen, Y, Jiang, J, He, J, Wang, Y, Zhang, K, Jiao, L, Li, L, Liu, F, Ma, W, Chen, Z, Fang, H, Zhang, W, Cong, R, Hegde, DD, Kalal, J, Akalwadi, N, Tabib, RA, Mudenagudi, U, Lin, YF, Lee, CM, Hsu, CC, Zhang, M, Qu, X, Liu, L, Liu, T, Chen, J, He, SY, Nathan, S, Uma, K, Sasithradevi, A, Bama, BS, Mansoor Roomi, SM, Benjdira, B, Ali, AM, Boulila, W, Dong, W, Li, Y, Hussein, A, Zhou, H, Chen, J, Xiao, Z & Li, Z 2025, NTIRE 2025 Challenge on Single Image Reflection Removal in the Wild: Datasets, Methods and Results. in Proceedings - 2025 IEEE/CVF Conference on Computer Vision and Pattern Recognition Workshops, CVPRW 2025. IEEE Computer Society Conference on Computer Vision and Pattern Recognition Workshops, IEEE Computer Society, pp. 1292-1302, 2025 IEEE/CVF Conference on Computer Vision and Pattern Recognition Workshops, CVPRW 2025, Nashville, United States, 11/06/25. https://doi.org/10.1109/CVPRW67362.2025.00122

NTIRE 2025 Challenge on Single Image Reflection Removal in the Wild: Datasets, Methods and Results. / Yang, Kangning; Cai, Jie; Ouyang, Ling et al.
Proceedings - 2025 IEEE/CVF Conference on Computer Vision and Pattern Recognition Workshops, CVPRW 2025. IEEE Computer Society, 2025. p. 1292-1302 (IEEE Computer Society Conference on Computer Vision and Pattern Recognition Workshops).

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution › peer-review

TY - GEN

T1 - NTIRE 2025 Challenge on Single Image Reflection Removal in the Wild

T2 - 2025 IEEE/CVF Conference on Computer Vision and Pattern Recognition Workshops, CVPRW 2025

AU - Yang, Kangning

AU - Cai, Jie

AU - Ouyang, Ling

AU - Vasluianu, Florin Alexandru

AU - Timofte, Radu

AU - Ding, Jiaming

AU - Sun, Huiming

AU - Fu, Lan

AU - Li, Jinlong

AU - Ho, Chiu Man

AU - Meng, Zibo

AU - Li, Mingjia

AU - Wang, Hainuo

AU - Hu, Qiming

AU - Wang, Jiarui

AU - Zhao, Hao

AU - Hu, Jin

AU - Guo, Xiaojie

AU - Yang, Mengru

AU - Jiang, Kui

AU - Guo, Jin

AU - Chen, Yiang

AU - Jiang, Junjun

AU - He, Jing

AU - Wang, Yiqing

AU - Zhang, Kexin

AU - Jiao, Licheng

AU - Li, Lingling

AU - Liu, Fang

AU - Ma, Wenping

AU - Chen, Zhiyang

AU - Fang, Hao

AU - Zhang, Wei

AU - Cong, Runmin

AU - Hegde, Dheeraj Damodhar

AU - Kalal, Jatin

AU - Akalwadi, Nikhil

AU - Tabib, Ramesh Ashok

AU - Mudenagudi, Uma

AU - Lin, Yu Fan

AU - Lee, Chia Ming

AU - Hsu, Chih Chung

AU - Zhang, Mengxin

AU - Qu, Xiaochao

AU - Liu, Luoqi

AU - Liu, Ting

AU - Chen, Jinshan

AU - He, Shu Yu

AU - Nathan, Sabari

AU - Uma, K.

AU - Sasithradevi, A.

AU - Bama, B. Sathya

AU - Mansoor Roomi, S. Mohamed

AU - Benjdira, Bilel

AU - Ali, Anas M.

AU - Boulila, Wadii

AU - Dong, Wei

AU - Li, Yunzhe

AU - Hussein, Ali

AU - Zhou, Han

AU - Chen, Jun

AU - Xiao, Zeyu

AU - Li, Zhuoyuan

PY - 2025

Y1 - 2025

N2 - In this paper, we review the NTIRE 2025 challenge on single-image reflection removal (SIRR) in the wild. SIRR is a fundamental task in image restoration. Despite progress in academic research, most methods are tested on synthetic images or limited real-world images, creating a gap in realworld applications. In this challenge, participants are required to process real-world images that cover a range of reflection scenarios and intensities, with the goal of generating clean images without reflections. The challenge attracted more than 200 registrations, with 11 of them participating in the final testing phase. The top-ranked methods advanced the state-of-the-art reflection removal performance and earned unanimous recognition from the five experts in the field. The proposed datasets are available at https://huggingface.co/datasets/qiuzhangTiTi/NTIRE2025-SIRR and the homepage of this challenge is at https://github.com/caijie0620/Reflection-Removal-in-thewild.

AB - In this paper, we review the NTIRE 2025 challenge on single-image reflection removal (SIRR) in the wild. SIRR is a fundamental task in image restoration. Despite progress in academic research, most methods are tested on synthetic images or limited real-world images, creating a gap in realworld applications. In this challenge, participants are required to process real-world images that cover a range of reflection scenarios and intensities, with the goal of generating clean images without reflections. The challenge attracted more than 200 registrations, with 11 of them participating in the final testing phase. The top-ranked methods advanced the state-of-the-art reflection removal performance and earned unanimous recognition from the five experts in the field. The proposed datasets are available at https://huggingface.co/datasets/qiuzhangTiTi/NTIRE2025-SIRR and the homepage of this challenge is at https://github.com/caijie0620/Reflection-Removal-in-thewild.

UR - https://www.scopus.com/pages/publications/105017844501

U2 - 10.1109/CVPRW67362.2025.00122

DO - 10.1109/CVPRW67362.2025.00122

M3 - 会议稿件

AN - SCOPUS:105017844501

T3 - IEEE Computer Society Conference on Computer Vision and Pattern Recognition Workshops

SP - 1292

EP - 1302

BT - Proceedings - 2025 IEEE/CVF Conference on Computer Vision and Pattern Recognition Workshops, CVPRW 2025

PB - IEEE Computer Society

Y2 - 11 June 2025 through 12 June 2025

ER -

Yang K, Cai J, Ouyang L, Vasluianu FA, Timofte R, Ding J et al. NTIRE 2025 Challenge on Single Image Reflection Removal in the Wild: Datasets, Methods and Results. In Proceedings - 2025 IEEE/CVF Conference on Computer Vision and Pattern Recognition Workshops, CVPRW 2025. IEEE Computer Society. 2025. p. 1292-1302. (IEEE Computer Society Conference on Computer Vision and Pattern Recognition Workshops). doi: 10.1109/CVPRW67362.2025.00122

NTIRE 2025 Challenge on Single Image Reflection Removal in the Wild: Datasets, Methods and Results

Abstract

Publication series

Conference

Access to Document

Other files and links

Fingerprint

Cite this