FerretNet: Efficient Synthetic Image Detection via Local Pixel Dependencies
Fuente:
arXiv
Salvato in:
| Autori principali: | , , , |
|---|---|
| Natura: | Preprint |
| Pubblicazione: |
2025
|
| Soggetti: | |
| Accesso online: | |
| Tags: |
Aggiungi Tag
Nessun Tag, puoi essere il primo ad aggiungerne!!
|
| _version_ | 1866911227484045312 |
|---|---|
| author | Liang, Shuqiao Liu, Jian Chen, Renzhang Guan, Quanlong |
| author_facet | Liang, Shuqiao Liu, Jian Chen, Renzhang Guan, Quanlong |
| contents | The increasing realism of synthetic images generated by advanced models such as VAEs, GANs, and LDMs poses significant challenges for synthetic image detection. To address this issue, we explore two artifact types introduced during the generation process: (1) latent distribution deviations and (2) decoding-induced smoothing effects, which manifest as inconsistencies in local textures, edges, and color transitions. Leveraging local pixel dependencies (LPD) properties rooted in Markov Random Fields, we reconstruct synthetic images using neighboring pixel information to expose disruptions in texture continuity and edge coherence. Building upon LPD, we propose FerretNet, a lightweight neural network with only 1.1M parameters that delivers efficient and robust synthetic image detection. Extensive experiments demonstrate that FerretNet, trained exclusively on the 4-class ProGAN dataset, achieves an average accuracy of 97.1% on an open-world benchmark comprising 22 generative models. Our code and datasets are publicly available at https://github.com/xigua7105/FerretNet. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2509_20890 |
| institution | arXiv |
| publishDate | 2025 |
| record_format | arxiv |
| spellingShingle | FerretNet: Efficient Synthetic Image Detection via Local Pixel Dependencies Liang, Shuqiao Liu, Jian Chen, Renzhang Guan, Quanlong Computer Vision and Pattern Recognition Artificial Intelligence I.5.1; I.5.2; I.2.10 The increasing realism of synthetic images generated by advanced models such as VAEs, GANs, and LDMs poses significant challenges for synthetic image detection. To address this issue, we explore two artifact types introduced during the generation process: (1) latent distribution deviations and (2) decoding-induced smoothing effects, which manifest as inconsistencies in local textures, edges, and color transitions. Leveraging local pixel dependencies (LPD) properties rooted in Markov Random Fields, we reconstruct synthetic images using neighboring pixel information to expose disruptions in texture continuity and edge coherence. Building upon LPD, we propose FerretNet, a lightweight neural network with only 1.1M parameters that delivers efficient and robust synthetic image detection. Extensive experiments demonstrate that FerretNet, trained exclusively on the 4-class ProGAN dataset, achieves an average accuracy of 97.1% on an open-world benchmark comprising 22 generative models. Our code and datasets are publicly available at https://github.com/xigua7105/FerretNet. |
| title | FerretNet: Efficient Synthetic Image Detection via Local Pixel Dependencies |
| topic | Computer Vision and Pattern Recognition Artificial Intelligence I.5.1; I.5.2; I.2.10 |
| url | https://arxiv.org/abs/2509.20890 |