-
[pdf]
[supp]
[bibtex]@InProceedings{Ershov_2025_CVPR, author = {Ershov, Egor and Korchagin, Sergey and Khalin, Alexei and Panshin, Artyom and Terekhin, Arseniy and Zaychenkova, Ekaterina and Lobarev, Georgiy and Plokhotnyuk, Vsevolod and Abramov, Denis and Zhdanov, Elisey and Dorogova, Sofia and Mamedov, Yasin and Banic, Nikola and Perevozchikov, Georgii and Timofte, Radu and Zhang, Lize and Zhang, Yuqian and Liu, Shuai and Feng, Chaoyu and Wang, Luyang and Huang, Yibin and Shao, Guangqi and Wang, Xiaotao and Lei, Lei and Pan, Sishun and Zhong, Zhiqiang and Yang, Yang and Ali, Anas M. and Aloqayli, Hamad and Benjdira, Bilel and Boulila, Wadii and Xiaoyang, Ma and Zijun, Gao and Leyi, Xing and He, Zongqi and Zuo, Yushen and Xiao, Zhe and Chan, Kin-Chung and Li, Hanmin and Xiao, Jun and Lam, Kin-Man and Wu, Yunpeng and Manzura, Dmitrij and Storonkin, Daniil and Guo, Weixin and Xu, Kele and Xu, Qisheng and Gao, Zijian and Wan, Tianjiao and Vampilov, Buda and Kinli, Furkan and Kirac, Furkan}, title = {NTIRE 2025 Challenge on Night Photography Rendering}, booktitle = {Proceedings of the Computer Vision and Pattern Recognition Conference (CVPR) Workshops}, month = {June}, year = {2025}, pages = {1514-1524} }
NTIRE 2025 Challenge on Night Photography Rendering
Abstract
This paper presents a review of the NTIRE 2025 challenge on night photography rendering. The goal of the challenge was to find solutions that process raw smartphone images taken in nighttime conditions, producing output images that replicate the pipeline of a professional camera. Unlike the previous year's competition, this year a unique imaging setup was introduced: each scene was captured both by a smartphone and a professional camera with the help of a beam-splitter cube. There were two nominations: best objective quality and mean opinion score-based quality. To evaluate the objective score, the SSIM and PSNR metrics were utilized. To evaluate the MOS-based quality, a sufficient number of viewers were asked to assess the similarity of the proposed solutions to the ground truth image taken by the professional camera. The top ranking participants' solutions effectively represent the state-of-the-art in nighttime photography rendering. More results can be found at https://nightimaging.org.
Related Material