NTIRE 2026 Challenge on Robust AI-Generated Image Detection in the Wild
Fuente:
arXiv
Salvato in:
| Autori principali: | , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , |
|---|---|
| Natura: | Preprint |
| Pubblicazione: |
2026
|
| Soggetti: | |
| Accesso online: | |
| Tags: |
Aggiungi Tag
Nessun Tag, puoi essere il primo ad aggiungerne!!
|
| _version_ | 1866918443053219840 |
|---|---|
| author | Gushchin, Aleksandr Abud, Khaled Shumitskaya, Ekaterina Filippov, Artem Bychkov, Georgii Lavrushkin, Sergey Erofeev, Mikhail Antsiferova, Anastasia Chen, Changsheng Tan, Shunquan Timofte, Radu Vatolin, Dmitry Song, Chuanbiao Yu, Zijian Tan, Hao Lan, Jun Yang, Zhiqiang Tang, Yongwei Wu, Zhiqiang Seow, Jia Wen Koay, Hong Vin Ren, Haodong Xu, Feng Chen, Shuai Xia, Ruiyang Zhang, Qi Xu, Yaowen Zou, Zhaofan Sun, Hao Lu, Dagong Yao, Mufeng Xu, Xinlei Wu, Fei Guo, Fengjun Luo, Cong Sharma, Hardik Negi, Aashish Shaily, Prateek Kumar, Jayant Chaudhary, Sachin Dudhane, Akshay Hambarde, Praful Shukla, Amit Tu, Zhilin Li, Fengpeng Zhang, Jiamin Fei, Jianwei Li, Kemou Wu, Haiwei Benjdira, Bilel Ali, Anas M. Boulila, Wadii Qu, Chenfan Li, Junchi |
| author_facet | Gushchin, Aleksandr Abud, Khaled Shumitskaya, Ekaterina Filippov, Artem Bychkov, Georgii Lavrushkin, Sergey Erofeev, Mikhail Antsiferova, Anastasia Chen, Changsheng Tan, Shunquan Timofte, Radu Vatolin, Dmitry Song, Chuanbiao Yu, Zijian Tan, Hao Lan, Jun Yang, Zhiqiang Tang, Yongwei Wu, Zhiqiang Seow, Jia Wen Koay, Hong Vin Ren, Haodong Xu, Feng Chen, Shuai Xia, Ruiyang Zhang, Qi Xu, Yaowen Zou, Zhaofan Sun, Hao Lu, Dagong Yao, Mufeng Xu, Xinlei Wu, Fei Guo, Fengjun Luo, Cong Sharma, Hardik Negi, Aashish Shaily, Prateek Kumar, Jayant Chaudhary, Sachin Dudhane, Akshay Hambarde, Praful Shukla, Amit Tu, Zhilin Li, Fengpeng Zhang, Jiamin Fei, Jianwei Li, Kemou Wu, Haiwei Benjdira, Bilel Ali, Anas M. Boulila, Wadii Qu, Chenfan Li, Junchi |
| contents | This paper presents an overview of the NTIRE 2026 Challenge on Robust AI-Generated Image Detection in the Wild, held in conjunction with the NTIRE workshop at CVPR 2026. The goal of this challenge was to develop detection models capable of distinguishing real images from generated ones in realistic scenarios: the images are often transformed (cropped, resized, compressed, blurred) for practical usage, and therefore, the detection models should be robust to such transformations. The challenge is based on a novel dataset consisting of 108,750 real and 185,750 AI-generated images from 42 generators comprising a large variety of open-source and closed-source models of various architectures, augmented with 36 image transformations. Methods were evaluated using ROC AUC on the full test set, including both transformed and untransformed images. A total of 511 participants registered, with 20 teams submitting valid final solutions. This report provides a comprehensive overview of the challenge, describes the proposed solutions, and can be used as a valuable reference for researchers and practitioners in increasing the robustness of the detection models to real-world transformations. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2604_11487 |
| institution | arXiv |
| publishDate | 2026 |
| record_format | arxiv |
| spellingShingle | NTIRE 2026 Challenge on Robust AI-Generated Image Detection in the Wild Gushchin, Aleksandr Abud, Khaled Shumitskaya, Ekaterina Filippov, Artem Bychkov, Georgii Lavrushkin, Sergey Erofeev, Mikhail Antsiferova, Anastasia Chen, Changsheng Tan, Shunquan Timofte, Radu Vatolin, Dmitry Song, Chuanbiao Yu, Zijian Tan, Hao Lan, Jun Yang, Zhiqiang Tang, Yongwei Wu, Zhiqiang Seow, Jia Wen Koay, Hong Vin Ren, Haodong Xu, Feng Chen, Shuai Xia, Ruiyang Zhang, Qi Xu, Yaowen Zou, Zhaofan Sun, Hao Lu, Dagong Yao, Mufeng Xu, Xinlei Wu, Fei Guo, Fengjun Luo, Cong Sharma, Hardik Negi, Aashish Shaily, Prateek Kumar, Jayant Chaudhary, Sachin Dudhane, Akshay Hambarde, Praful Shukla, Amit Tu, Zhilin Li, Fengpeng Zhang, Jiamin Fei, Jianwei Li, Kemou Wu, Haiwei Benjdira, Bilel Ali, Anas M. Boulila, Wadii Qu, Chenfan Li, Junchi Computer Vision and Pattern Recognition This paper presents an overview of the NTIRE 2026 Challenge on Robust AI-Generated Image Detection in the Wild, held in conjunction with the NTIRE workshop at CVPR 2026. The goal of this challenge was to develop detection models capable of distinguishing real images from generated ones in realistic scenarios: the images are often transformed (cropped, resized, compressed, blurred) for practical usage, and therefore, the detection models should be robust to such transformations. The challenge is based on a novel dataset consisting of 108,750 real and 185,750 AI-generated images from 42 generators comprising a large variety of open-source and closed-source models of various architectures, augmented with 36 image transformations. Methods were evaluated using ROC AUC on the full test set, including both transformed and untransformed images. A total of 511 participants registered, with 20 teams submitting valid final solutions. This report provides a comprehensive overview of the challenge, describes the proposed solutions, and can be used as a valuable reference for researchers and practitioners in increasing the robustness of the detection models to real-world transformations. |
| title | NTIRE 2026 Challenge on Robust AI-Generated Image Detection in the Wild |
| topic | Computer Vision and Pattern Recognition |
| url | https://arxiv.org/abs/2604.11487 |