STD-FD: Spatio-Temporal Distribution Fitting Deviation for AIGC Forgery Identification

Hengrui Lou, Zunlei Feng, Jinsong Geng, Erteng Liu, Jie Lei, Lechao Cheng, Jie Song, Mingli Song, Yijun Bei
Proceedings of the 42nd International Conference on Machine Learning, PMLR 267:40328-40343, 2025.

Abstract

With the rise of AIGC technologies, particularly diffusion models, highly realistic fake images that can deceive human visual perception has become feasible. Consequently, various forgery detection methods have emerged. However, existing methods treat the generation process of fake images as either a black-box or an auxiliary tool, offering limited insights into its underlying mechanisms. In this paper, we propose Spatio-Temporal Distribution Fitting Deviation (STD-FD) for AIGC forgery detection, which explores the generative process in detail. By decomposing and reconstructing data within generative diffusion models, initial experiments reveal temporal distribution fitting deviations during the image reconstruction process. These deviations are captured through reconstruction noise maps for each spatial semantic unit, derived via a super-resolution algorithm. Critical discriminative patterns, termed DFactors, are identified through statistical modeling of these deviations. Extensive experiments show that STD-FD effectively captures distribution patterns in AIGC-generated data, demonstrating strong robustness and generalizability while outperforming state-of-the-art (SOTA) methods on major datasets. The source code is available at this link.

Cite this Paper

BibTeX
@InProceedings{pmlr-v267-lou25a, title = {{STD}-{FD}: Spatio-Temporal Distribution Fitting Deviation for {AIGC} Forgery Identification}, author = {Lou, Hengrui and Feng, Zunlei and Geng, Jinsong and Liu, Erteng and Lei, Jie and Cheng, Lechao and Song, Jie and Song, Mingli and Bei, Yijun}, booktitle = {Proceedings of the 42nd International Conference on Machine Learning}, pages = {40328--40343}, year = {2025}, editor = {Singh, Aarti and Fazel, Maryam and Hsu, Daniel and Lacoste-Julien, Simon and Berkenkamp, Felix and Maharaj, Tegan and Wagstaff, Kiri and Zhu, Jerry}, volume = {267}, series = {Proceedings of Machine Learning Research}, month = {13--19 Jul}, publisher = {PMLR}, pdf = {https://raw.githubusercontent.com/mlresearch/v267/main/assets/lou25a/lou25a.pdf}, url = {https://proceedings.mlr.press/v267/lou25a.html}, abstract = {With the rise of AIGC technologies, particularly diffusion models, highly realistic fake images that can deceive human visual perception has become feasible. Consequently, various forgery detection methods have emerged. However, existing methods treat the generation process of fake images as either a black-box or an auxiliary tool, offering limited insights into its underlying mechanisms. In this paper, we propose Spatio-Temporal Distribution Fitting Deviation (STD-FD) for AIGC forgery detection, which explores the generative process in detail. By decomposing and reconstructing data within generative diffusion models, initial experiments reveal temporal distribution fitting deviations during the image reconstruction process. These deviations are captured through reconstruction noise maps for each spatial semantic unit, derived via a super-resolution algorithm. Critical discriminative patterns, termed DFactors, are identified through statistical modeling of these deviations. Extensive experiments show that STD-FD effectively captures distribution patterns in AIGC-generated data, demonstrating strong robustness and generalizability while outperforming state-of-the-art (SOTA) methods on major datasets. The source code is available at this link.} }
Endnote
%0 Conference Paper %T STD-FD: Spatio-Temporal Distribution Fitting Deviation for AIGC Forgery Identification %A Hengrui Lou %A Zunlei Feng %A Jinsong Geng %A Erteng Liu %A Jie Lei %A Lechao Cheng %A Jie Song %A Mingli Song %A Yijun Bei %B Proceedings of the 42nd International Conference on Machine Learning %C Proceedings of Machine Learning Research %D 2025 %E Aarti Singh %E Maryam Fazel %E Daniel Hsu %E Simon Lacoste-Julien %E Felix Berkenkamp %E Tegan Maharaj %E Kiri Wagstaff %E Jerry Zhu %F pmlr-v267-lou25a %I PMLR %P 40328--40343 %U https://proceedings.mlr.press/v267/lou25a.html %V 267 %X With the rise of AIGC technologies, particularly diffusion models, highly realistic fake images that can deceive human visual perception has become feasible. Consequently, various forgery detection methods have emerged. However, existing methods treat the generation process of fake images as either a black-box or an auxiliary tool, offering limited insights into its underlying mechanisms. In this paper, we propose Spatio-Temporal Distribution Fitting Deviation (STD-FD) for AIGC forgery detection, which explores the generative process in detail. By decomposing and reconstructing data within generative diffusion models, initial experiments reveal temporal distribution fitting deviations during the image reconstruction process. These deviations are captured through reconstruction noise maps for each spatial semantic unit, derived via a super-resolution algorithm. Critical discriminative patterns, termed DFactors, are identified through statistical modeling of these deviations. Extensive experiments show that STD-FD effectively captures distribution patterns in AIGC-generated data, demonstrating strong robustness and generalizability while outperforming state-of-the-art (SOTA) methods on major datasets. The source code is available at this link.
APA
Lou, H., Feng, Z., Geng, J., Liu, E., Lei, J., Cheng, L., Song, J., Song, M. & Bei, Y.. (2025). STD-FD: Spatio-Temporal Distribution Fitting Deviation for AIGC Forgery Identification. Proceedings of the 42nd International Conference on Machine Learning, in Proceedings of Machine Learning Research 267:40328-40343 Available from https://proceedings.mlr.press/v267/lou25a.html.

Related Material