No Pixel Left Behind: A Detail-Preserving Architecture for Robust High-Resolution AI-Generated Image Detection
Fuente:
arXiv
Saved in:
| Main Authors: | , , , , , , , , , |
|---|---|
| Format: | Preprint |
| Published: |
2025
|
| Subjects: | |
| Online Access: | |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1866908500278378496 |
|---|---|
| author | Mu, Lianrui Xingze, Zou Bai, Jianhong Hu, Jiaqi Zheng, Wenjie Ye, Jiangnan Zhuang, Jiedong Ali, Mudassar Wang, Jing Hu, Haoji |
| author_facet | Mu, Lianrui Xingze, Zou Bai, Jianhong Hu, Jiaqi Zheng, Wenjie Ye, Jiangnan Zhuang, Jiedong Ali, Mudassar Wang, Jing Hu, Haoji |
| contents | The rapid growth of high-resolution, meticulously crafted AI-generated images poses a significant challenge to existing detection methods, which are often trained and evaluated on low-resolution, automatically generated datasets that do not align with the complexities of high-resolution scenarios. A common practice is to resize or center-crop high-resolution images to fit standard network inputs. However, without full coverage of all pixels, such strategies risk either obscuring subtle, high-frequency artifacts or discarding information from uncovered regions, leading to input information loss. In this paper, we introduce the High-Resolution Detail-Aggregation Network (HiDA-Net), a novel framework that ensures no pixel is left behind. We use the Feature Aggregation Module (FAM), which fuses features from multiple full-resolution local tiles with a down-sampled global view of the image. These local features are aggregated and fused with global representations for final prediction, ensuring that native-resolution details are preserved and utilized for detection. To enhance robustness against challenges such as localized AI manipulations and compression, we introduce Token-wise Forgery Localization (TFL) module for fine-grained spatial sensitivity and JPEG Quality Factor Estimation (QFE) module to disentangle generative artifacts from compression noise explicitly. Furthermore, to facilitate future research, we introduce HiRes-50K, a new challenging benchmark consisting of 50,568 images with up to 64 megapixels. Extensive experiments show that HiDA-Net achieves state-of-the-art, increasing accuracy by over 13% on the challenging Chameleon dataset and 10% on our HiRes-50K. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2508_17346 |
| institution | arXiv |
| publishDate | 2025 |
| record_format | arxiv |
| spellingShingle | No Pixel Left Behind: A Detail-Preserving Architecture for Robust High-Resolution AI-Generated Image Detection Mu, Lianrui Xingze, Zou Bai, Jianhong Hu, Jiaqi Zheng, Wenjie Ye, Jiangnan Zhuang, Jiedong Ali, Mudassar Wang, Jing Hu, Haoji Computer Vision and Pattern Recognition The rapid growth of high-resolution, meticulously crafted AI-generated images poses a significant challenge to existing detection methods, which are often trained and evaluated on low-resolution, automatically generated datasets that do not align with the complexities of high-resolution scenarios. A common practice is to resize or center-crop high-resolution images to fit standard network inputs. However, without full coverage of all pixels, such strategies risk either obscuring subtle, high-frequency artifacts or discarding information from uncovered regions, leading to input information loss. In this paper, we introduce the High-Resolution Detail-Aggregation Network (HiDA-Net), a novel framework that ensures no pixel is left behind. We use the Feature Aggregation Module (FAM), which fuses features from multiple full-resolution local tiles with a down-sampled global view of the image. These local features are aggregated and fused with global representations for final prediction, ensuring that native-resolution details are preserved and utilized for detection. To enhance robustness against challenges such as localized AI manipulations and compression, we introduce Token-wise Forgery Localization (TFL) module for fine-grained spatial sensitivity and JPEG Quality Factor Estimation (QFE) module to disentangle generative artifacts from compression noise explicitly. Furthermore, to facilitate future research, we introduce HiRes-50K, a new challenging benchmark consisting of 50,568 images with up to 64 megapixels. Extensive experiments show that HiDA-Net achieves state-of-the-art, increasing accuracy by over 13% on the challenging Chameleon dataset and 10% on our HiRes-50K. |
| title | No Pixel Left Behind: A Detail-Preserving Architecture for Robust High-Resolution AI-Generated Image Detection |
| topic | Computer Vision and Pattern Recognition |
| url | https://arxiv.org/abs/2508.17346 |