The Fourth Challenge on Image Super-Resolution ($\times$4) at NTIRE 2026: Benchmark Results and Method Overview
Abstract
This paper presents the NTIRE 2026 image super-resolution (4) challenge, one of the associated competitions of the NTIRE 2026 Workshop at CVPR 2026. The challenge aims to reconstruct high-resolution (HR) images from low-resolution (LR) inputs generated through bicubic downsampling with a 4 scaling factor. The objective is to develop effective super-resolution solutions and analyze recent advances in the field. To reflect the evolving objectives of image super-resolution, the challenge includes two tracks: (1) a restoration track, which emphasizes pixel-wise fidelity and ranks submissions based on PSNR; and (2) a perceptual track, which focuses on visual realism and evaluates results using a perceptual score. A total of 194 participants registered for the challenge, with 31 teams submitting valid entries. This report summarizes the challenge design, datasets, evaluation protocol, main results, and methods of participating teams. The challenge provides a unified benchmark and offers insights into current progress and future directions in image super-resolution.
Keywords
Cite
@article{arxiv.2604.14558,
title = {The Fourth Challenge on Image Super-Resolution ($\times$4) at NTIRE 2026: Benchmark Results and Method Overview},
author = {Zheng Chen and Kai Liu and Jingkai Wang and Xianglong Yan and Jianze Li and Ziqing Zhang and Jue Gong and Jiatong Li and Lei Sun and Xiaoyang Liu and Radu Timofte and Yulun Zhang and Jihye Park and Yoonjin Im and Hyungju Chun and Hyunhee Park and MinKyu Park and Zheng Xie and Xiangyu Kong and Weijun Yuan and Zhan Li and Qiurong Song and Luen Zhu and Fengkai Zhang and Xinzhe Zhu and Junyang Chen and Congyu Wang and Yixin Yang and Zhaorun Zhou and Jiangxin Dong and Jinshan Pan and Shengwei Wang and Jiajie Ou and Baiang Li and Sizhuo Ma and Qiang Gao and Jusheng Zhang and Jian Wang and Keze Wang and Yijiao Liu and Yingsi Chen and Hui Li and Yu Wang and Congchao Zhu and Saeed Ahmad and Ik Hyun Lee and Jun Young Park and Ji Hwan Yoon and Kainan Yan and Zian Wang and Weibo Wang and Shihao Zou and Chao Dong and Wei Zhou and Linfeng Li and Jaeseong Lee and Jaeho Chae and Jinwoo Kim and Seonjoo Kim and Yucong Hong and Zhenming Yan and Junye Chen and Ruize Han and Song Wang and Yuxuan Jiang and Chengxi Zeng and Tianhao Peng and Fan Zhang and David Bull and Tongyao Mu and Qiong Cao and Yifan Wang and Youwei Pan and Leilei Cao and Xiaoping Peng and Wei Deng and Yifei Chen and Wenbo Xiong and Xian Hu and Yuxin Zhang and Xiaoyun Cheng and Yang Ji and Zonghao Chen and Zhihao Xue and Junqin Hu and Nihal Kumar and Snehal Singh Tomar and Klaus Mueller and Surya Vashisth and Prateek Shaily and Jayant Kumar and Hardik Sharma and Ashish Negi and Sachin Chaudhary and Akshay Dudhane and Praful Hambarde and Amit Shukla and Shijun Shi and Jiangning Zhang and Yong Liu and Kai Hu and Jing Xu and Xianfang Zeng and Amitesh M and Hariharan S and Chia-Ming Lee and Yu-Fan Lin and Chih-Chung Hsu and Nishalini K and Sreenath K A and Bilel Benjdira and Anas M. Ali and Wadii Boulila and Shuling Zheng and Zhiheng Fu and Feng Zhang and Zhanglu Chen and Boyang Yao and Nikhil Pathak and Aagam Jain and Milan Kumar and Kishor Upla and Vivek Chavda and Sarang N S and Raghavendra Ramachandra and Zhipeng Zhang and Qi Wang and Shiyu Wang and Jiachen Tu and Guoyi Xu and Yaoxin Jiang and Jiajia Liu and Yaokun Shi and Yuqi Li and Chuanguang Yang and Weilun Feng and Zhuzhi Hong and Hao Wu and Junming Liu and Yingli Tian and Amish Bhushan Kulkarni and Tejas R R Shet and Saakshi M Vernekar and Nikhil Akalwadi and Kaushik Mallibhat and Ramesh Ashok Tabib and Uma Mudenagudi and Yuwen Pan and Tianrun Chen and Deyi Ji and Qi Zhu and Lanyun Zhu and Heyan Zhangyi},
journal= {arXiv preprint arXiv:2604.14558},
year = {2026}
}
Comments
NTIRE 2026 webpage: https://cvlai.net/ntire/2026. Code: https://github.com/zhengchen1999/NTIRE2026_ImageSR_x4