NTIRE 2025 challenge on Text to Image Generation Model Quality Assessment
This paper reports on the NTIRE 2025 challenge on Text to Image (T2I) generation model quality assessment, which will be held in conjunction with the New Trends in Image Restoration and Enhancement Workshop (NTIRE) at CVPR 2025. The aim of this challenge is to address the fine-grained quality assessment of text-to-image generation models. This challenge evaluates text-to-image models from two aspects: image-text alignment and image structural distortion detection, and is divided into the alignment track and the structural track. The alignment track uses the EvalMuse-40K, which contains around 40K AI-Generated Images (AIGIs) generated by 20 popular generative models. The alignment track has a total of 371 registered participants. A total of 1,883 submissions are received in the development phase, and 507 submissions are received in the test phase. Finally, 12 participating teams submitted their models and fact sheets. The structure track uses the EvalMuse-Structure, which contains 10,000 AI-Generated Images (AIGIs) with corresponding structural distortion mask. A total of 211 participants have registered in the structure track. A total of 1155 submissions are received in the development phase, and 487 submissions are received in the test phase. Finally, 8 participating teams submitted their models and fact sheets. Almost all methods have achieved better results than baseline methods, and the winning methods in both tracks have demonstrated superior prediction performance on T2I model quality assessment.
View on arXiv@article{han2025_2505.16314,
title={ NTIRE 2025 challenge on Text to Image Generation Model Quality Assessment },
author={ Shuhao Han and Haotian Fan and Fangyuan Kong and Wenjie Liao and Chunle Guo and Chongyi Li and Radu Timofte and Liang Li and Tao Li and Junhui Cui and Yunqiu Wang and Yang Tai and Jingwei Sun and Jianhui Sun and Xinli Yue and Tianyi Wang and Huan Hou and Junda Lu and Xinyang Huang and Zitang Zhou and Zijian Zhang and Xuhui Zheng and Xuecheng Wu and Chong Peng and Xuezhi Cao and Trong-Hieu Nguyen-Mau and Minh-Hoang Le and Minh-Khoa Le-Phan and Duy-Nam Ly and Hai-Dang Nguyen and Minh-Triet Tran and Yukang Lin and Yan Hong and Chuanbiao Song and Siyuan Li and Jun Lan and Zhichao Zhang and Xinyue Li and Wei Sun and Zicheng Zhang and Yunhao Li and Xiaohong Liu and Guangtao Zhai and Zitong Xu and Huiyu Duan and Jiarui Wang and Guangji Ma and Liu Yang and Lu Liu and Qiang Hu and Xiongkuo Min and Zichuan Wang and Zhenchen Tang and Bo Peng and Jing Dong and Fengbin Guan and Zihao Yu and Yiting Lu and Wei Luo and Xin Li and Minhao Lin and Haofeng Chen and Xuanxuan He and Kele Xu and Qisheng Xu and Zijian Gao and Tianjiao Wan and Bo-Cheng Qiu and Chih-Chung Hsu and Chia-ming Lee and Yu-Fan Lin and Bo Yu and Zehao Wang and Da Mu and Mingxiu Chen and Junkang Fang and Huamei Sun and Wending Zhao and Zhiyu Wang and Wang Liu and Weikang Yu and Puhong Duan and Bin Sun and Xudong Kang and Shutao Li and Shuai He and Lingzhi Fu and Heng Cong and Rongyu Zhang and Jiarong He and Zhishan Qiao and Yongqing Huang and Zewen Chen and Zhe Pang and Juan Wang and Jian Guo and Zhizhuo Shao and Ziyu Feng and Bing Li and Weiming Hu },
journal={arXiv preprint arXiv:2505.16314},
year={ 2025 }
}