-
[pdf]
[bibtex]@InProceedings{Chen_2024_CVPR, author = {Chen, Zheng and Wu, Zongwei and Zamfir, Eduard and Zhang, Kai and Zhang, Yulun and Timofte, Radu and Yang, Xiaokang and Yu, Hongyuan and Wan, Cheng and Hong, Yuxin and Huang, Zhijuan and Zou, Yajun and Huang, Yuan and Lin, Jiamin and Han, Bingnan and Guan, Xianyu and Yu, Yongsheng and Zhang, Daoan and Yin, Xuanwu and Zuo, Kunlong and Hao, Jinhua and Zhao, Kai and Yuan, Kun and Sun, Ming and Zhou, Chao and An, Hongyu and Zhang, Xinfeng and Song, Zhiyuan and Dong, Ziyue and Zhao, Qing and Xu, Xiaogang and Wei, Pengxu and Dou, Zhi-Chao and Wang, Gui-Ling and Hsu, Chih-Chung and Lee, Chia-Ming and Chou, Yi-Shiuan and Korkmaz, Cansu and Tekalp, A. Murat and Wei, Yubin and Yan, Xiaole and Li, Binren and Chen, Haonan and Zhang, Siqi and Chen, Sihan and Joshi, Amogh and Akalwadi, Nikhil and Malagi, Sampada and Yashaswini, Palani and Desai, Chaitra and Tabib, Ramesh Ashok and Patil, Ujwala and Mudenagudi, Uma and Sarvaiya, Anjali and Choksy, Pooja and Joshi, Jagrit and Kawa, Shubh and Upla, Kishor and Patwardhan, Sushrut and Ramachandra, Raghavendra and Hossain, Sadat and Park, Geongi and Uddin, S.M. Nadim and Xu, Hao and Guo, Yanhui and Urumbekov, Aman and Yan, Xingzhuo and Hao, Wei and Fu, Minghan and Orais, Isaac and Smith, Samuel and Liu, Ying and Jia, Wangwang and Xu, Qisheng and Xu, Kele and Yuan, Weijun and Li, Zhan and Kuang, Wenqin and Guan, Ruijin and Deng, Ruting and Zhang, Zhao and Wang, Bo and Zhao, Suiyi and Luo, Yan and Wei, Yanyan and Khan, Asif Hussain and Micheloni, Christian and Martinel, Niki}, title = {NTIRE 2024 Challenge on Image Super-Resolution (x4): Methods and Results}, booktitle = {Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR) Workshops}, month = {June}, year = {2024}, pages = {6108-6132} }
NTIRE 2024 Challenge on Image Super-Resolution (x4): Methods and Results
Abstract
This paper reviews the NTIRE 2024 challenge on image super-resolution (x4) highlighting the solutions proposed and the outcomes obtained. The challenge involves generating corresponding high-resolution (HR) images magnified by a factor of four from low-resolution (LR) inputs using prior information. The LR images originate from bicubic downsampling degradation. The aim of the challenge is to obtain designs/solutions with the most advanced SR performance with no constraints on computational resources (e.g. model size and FLOPs) or training data. The track of this challenge assesses performance with the PSNR metric on the DIV2K testing dataset. The competition attracted 199 registrants with 20 teams submitting valid entries. This collective endeavour not only pushes the boundaries of performance in single-image SR but also offers a comprehensive overview of current trends in this field.
Related Material