Saved in:
| Main Authors: | , , , |
|---|---|
| Format: | Preprint |
| Published: |
2024
|
| Subjects: | |
| Online Access: | https://arxiv.org/abs/2406.00856 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1866914822940000256 |
|---|---|
| author | Lim, Yewon Lee, Changyeon Kim, Aerin Etzioni, Oren |
| author_facet | Lim, Yewon Lee, Changyeon Kim, Aerin Etzioni, Oren |
| contents | A dramatic influx of diffusion-generated images has marked recent years, posing unique challenges to current detection technologies. While the task of identifying these images falls under binary classification, a seemingly straightforward category, the computational load is significant when employing the "reconstruction then compare" technique. This approach, known as DIRE (Diffusion Reconstruction Error), not only identifies diffusion-generated images but also detects those produced by GANs, highlighting the technique's broad applicability. To address the computational challenges and improve efficiency, we propose distilling the knowledge embedded in diffusion models to develop rapid deepfake detection models. Our approach, aimed at creating a small, fast, cheap, and lightweight diffusion synthesized deepfake detector, maintains robust performance while significantly reducing operational demands. Maintaining performance, our experimental results indicate an inference speed 3.2 times faster than the existing DIRE framework. This advance not only enhances the practicality of deploying these systems in real-world settings but also paves the way for future research endeavors that seek to leverage diffusion model knowledge. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2406_00856 |
| institution | arXiv |
| publishDate | 2024 |
| record_format | arxiv |
| spellingShingle | DistilDIRE: A Small, Fast, Cheap and Lightweight Diffusion Synthesized Deepfake Detection Lim, Yewon Lee, Changyeon Kim, Aerin Etzioni, Oren Computer Vision and Pattern Recognition Cryptography and Security Machine Learning A dramatic influx of diffusion-generated images has marked recent years, posing unique challenges to current detection technologies. While the task of identifying these images falls under binary classification, a seemingly straightforward category, the computational load is significant when employing the "reconstruction then compare" technique. This approach, known as DIRE (Diffusion Reconstruction Error), not only identifies diffusion-generated images but also detects those produced by GANs, highlighting the technique's broad applicability. To address the computational challenges and improve efficiency, we propose distilling the knowledge embedded in diffusion models to develop rapid deepfake detection models. Our approach, aimed at creating a small, fast, cheap, and lightweight diffusion synthesized deepfake detector, maintains robust performance while significantly reducing operational demands. Maintaining performance, our experimental results indicate an inference speed 3.2 times faster than the existing DIRE framework. This advance not only enhances the practicality of deploying these systems in real-world settings but also paves the way for future research endeavors that seek to leverage diffusion model knowledge. |
| title | DistilDIRE: A Small, Fast, Cheap and Lightweight Diffusion Synthesized Deepfake Detection |
| topic | Computer Vision and Pattern Recognition Cryptography and Security Machine Learning |
| url | https://arxiv.org/abs/2406.00856 |