-
[pdf]
[bibtex]@InProceedings{Perez-Pellitero_2022_CVPR, author = {P\'erez-Pellitero, Eduardo and Catley-Chandar, Sibi and Shaw, Richard and Leonardis, Ale\v{s} and Timofte, Radu and Zhang, Zexin and Liu, Cen and Peng, Yunbo and Lin, Yue and Yu, Gaocheng and Zhang, Jin and Ma, Zhe and Wang, Hongbin and Chen, Xiangyu and Wang, Xintao and Wu, Haiwei and Liu, Lin and Dong, Chao and Zhou, Jiantao and Yan, Qingsen and Zhang, Song and Chen, Weiye and Liu, Yuhang and Zhang, Zhen and Zhang, Yanning and Shi, Javen Qinfeng and Gong, Dong and Zhu, Dan and Sun, Mengdi and Chen, Guannan and Hu, Yang and Li, Haowei and Zou, Baozhu and Liu, Zhen and Lin, Wenjie and Jiang, Ting and Jiang, Chengzhi and Li, Xinpeng and Han, Mingyan and Fan, Haoqiang and Sun, Jian and Liu, Shuaicheng and Mar{\'\i}n-Vega, Juan and Sloth, Michael and Schneider-Kamp, Peter and R\"ottger, Richard and Li, Chunyang and Bao, Long and He, Gang and Xu, Ziyao and Xu, Li and Zhan, Gen and Sun, Ming and Wen, Xing and Li, Junlin and Li, Jinjing and Li, Jinjing and Li, Chenghua and Li, Chenghua and Gang, Ruipeng and Gang, Ruipeng and Li, Fangya and Li, Fangya and Liu, Chenming and Liu, Chenming and Feng, Shuang and Lei, Fei and Liu, Rui and Ruan, Junxiang and Dai, Tianhong and Li, Wei and Lu, Zhan and Liu, Hengyan and Huang, Peian and Ren, Guangyu and Luo, Yonglin and Liu, Chang and Tu, Qiang and Ma, Sai and Cao, Yizhen and Tel, Steven and Heyrman, Barthelemy and Ginhac, Dominique and Lee, Chul and Kim, Gahyeon and Park, Seonghyun and Vien, An Gia and Mai, Truong Thanh Nhat and Yoon, Howoon and Vo, Tu and Holston, Alexander and Zaheer, Sheir and Park, Chan Y.}, title = {NTIRE 2022 Challenge on High Dynamic Range Imaging: Methods and Results}, booktitle = {Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR) Workshops}, month = {June}, year = {2022}, pages = {1009-1023} }
NTIRE 2022 Challenge on High Dynamic Range Imaging: Methods and Results
Abstract
This paper reviews the challenge on constrained high dynamic range (HDR) imaging that was part of the New Trends in Image Restoration and Enhancement (NTIRE) workshop, held in conjunction with CVPR 2022. This manuscript focuses on the competition set-up, datasets, the proposed methods and their results. The challenge aims at estimating an HDR image from multiple respective low dynamic range (LDR) observations, which might suffer from under- or over-exposed regions and different sources of noise. The challenge is composed of two tracks with an emphasis on fidelity and complexity constraints: In Track 1, participants are asked to optimize objective fidelity scores while imposing a low-complexity constraint (i.e. solutions can not exceed a given number of operations). In Track 2, participants are asked to minimize the complexity of their solutions while imposing a constraint on fidelity scores (i.e. solutions are required to obtain a higher fidelity score than the prescribed baseline). Both tracks use the same data and metrics: Fidelity is measured by means of PSNR with respect to a ground-truth HDR image (computed both directly and with a canonical tonemapping operation), while complexity metrics include the number of operations (measured in GMACs) and runtime (in seconds).
Related Material