Saved in:
| Main Authors: | , , , |
|---|---|
| Format: | Preprint |
| Published: |
2026
|
| Subjects: | |
| Online Access: | https://arxiv.org/abs/2604.16083 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1866914483357614080 |
|---|---|
| author | Yu, Jieming Feng, Qiuxiao Wang, Zhuohan Ma, Xiaochen |
| author_facet | Yu, Jieming Feng, Qiuxiao Wang, Zhuohan Ma, Xiaochen |
| contents | With the rapid advancement of deep generative models, realistic fake images have become increasingly accessible, yet existing localization methods rely on complex designs and still struggle to generalize across manipulation types and imaging conditions. We present a simple but strong baseline based on DINOv3 with LoRA adaptation and a lightweight convolutional decoder. Under the CAT-Net protocol, our best model improves average pixel-level F1 by 17.0 points over the previous state of the art on four standard benchmarks using only 9.1\,M trainable parameters on top of a frozen ViT-L backbone, and even our smallest variant surpasses all prior specialized methods. LoRA consistently outperforms full fine-tuning across all backbone scales. Under the data-scarce MVSS-Net protocol, LoRA reaches an average F1 of 0.774 versus 0.530 for the strongest prior method, while full fine-tuning becomes highly unstable, suggesting that pre-trained representations encode forensic information that is better preserved than overwritten. The baseline also exhibits strong robustness to Gaussian noise, JPEG re-compression, and Gaussian blur. We hope this work can serve as a reliable baseline for the research community and a practical starting point for future image-forensic applications. Code is available at https://github.com/Irennnne/DINOv3-IML. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2604_16083 |
| institution | arXiv |
| publishDate | 2026 |
| record_format | arxiv |
| spellingShingle | DINOv3 Beats Specialized Detectors: A Simple Foundation Model Baseline for Image Forensics Yu, Jieming Feng, Qiuxiao Wang, Zhuohan Ma, Xiaochen Computer Vision and Pattern Recognition With the rapid advancement of deep generative models, realistic fake images have become increasingly accessible, yet existing localization methods rely on complex designs and still struggle to generalize across manipulation types and imaging conditions. We present a simple but strong baseline based on DINOv3 with LoRA adaptation and a lightweight convolutional decoder. Under the CAT-Net protocol, our best model improves average pixel-level F1 by 17.0 points over the previous state of the art on four standard benchmarks using only 9.1\,M trainable parameters on top of a frozen ViT-L backbone, and even our smallest variant surpasses all prior specialized methods. LoRA consistently outperforms full fine-tuning across all backbone scales. Under the data-scarce MVSS-Net protocol, LoRA reaches an average F1 of 0.774 versus 0.530 for the strongest prior method, while full fine-tuning becomes highly unstable, suggesting that pre-trained representations encode forensic information that is better preserved than overwritten. The baseline also exhibits strong robustness to Gaussian noise, JPEG re-compression, and Gaussian blur. We hope this work can serve as a reliable baseline for the research community and a practical starting point for future image-forensic applications. Code is available at https://github.com/Irennnne/DINOv3-IML. |
| title | DINOv3 Beats Specialized Detectors: A Simple Foundation Model Baseline for Image Forensics |
| topic | Computer Vision and Pattern Recognition |
| url | https://arxiv.org/abs/2604.16083 |