Perceptual Classifiers: Detecting Generative Images using Perceptual Features
Fuente:
arXiv
Guardado en:
| Autores principales: | , , , |
|---|---|
| Formato: | Preprint |
| Publicado: |
2025
|
| Materias: | |
| Acceso en línea: | |
| Etiquetas: |
Agregar Etiqueta
Sin Etiquetas, Sea el primero en etiquetar este registro!
|
| _version_ | 1866915405952450560 |
|---|---|
| author | Durbha, Krishna Srikar Venkataramanan, Asvin Kumar Sureddi, Rajesh Bovik, Alan C. |
| author_facet | Durbha, Krishna Srikar Venkataramanan, Asvin Kumar Sureddi, Rajesh Bovik, Alan C. |
| contents | Image Quality Assessment (IQA) models are employed in many practical image and video processing pipelines to reduce storage, minimize transmission costs, and improve the Quality of Experience (QoE) of millions of viewers. These models are sensitive to a diverse range of image distortions and can accurately predict image quality as judged by human viewers. Recent advancements in generative models have resulted in a significant influx of "GenAI" content on the internet. Existing methods for detecting GenAI content have progressed significantly with improved generalization performance on images from unseen generative models. Here, we leverage the capabilities of existing IQA models, which effectively capture the manifold of real images within a bandpass statistical space, to distinguish between real and AI-generated images. We investigate the generalization ability of these perceptual classifiers to the task of GenAI image detection and evaluate their robustness against various image degradations. Our results show that a two-layer network trained on the feature space of IQA models demonstrates state-of-the-art performance in detecting fake images across generative models, while maintaining significant robustness against image degradations. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2507_17240 |
| institution | arXiv |
| publishDate | 2025 |
| record_format | arxiv |
| spellingShingle | Perceptual Classifiers: Detecting Generative Images using Perceptual Features Durbha, Krishna Srikar Venkataramanan, Asvin Kumar Sureddi, Rajesh Bovik, Alan C. Computer Vision and Pattern Recognition Image Quality Assessment (IQA) models are employed in many practical image and video processing pipelines to reduce storage, minimize transmission costs, and improve the Quality of Experience (QoE) of millions of viewers. These models are sensitive to a diverse range of image distortions and can accurately predict image quality as judged by human viewers. Recent advancements in generative models have resulted in a significant influx of "GenAI" content on the internet. Existing methods for detecting GenAI content have progressed significantly with improved generalization performance on images from unseen generative models. Here, we leverage the capabilities of existing IQA models, which effectively capture the manifold of real images within a bandpass statistical space, to distinguish between real and AI-generated images. We investigate the generalization ability of these perceptual classifiers to the task of GenAI image detection and evaluate their robustness against various image degradations. Our results show that a two-layer network trained on the feature space of IQA models demonstrates state-of-the-art performance in detecting fake images across generative models, while maintaining significant robustness against image degradations. |
| title | Perceptual Classifiers: Detecting Generative Images using Perceptual Features |
| topic | Computer Vision and Pattern Recognition |
| url | https://arxiv.org/abs/2507.17240 |