Data-Independent Operator: A Training-Free Artifact Representation Extractor for Generalizable Deepfake Detection
Fuente:
arXiv
Guardado en:
| Autores principales: | , , , , , , |
|---|---|
| Formato: | Preprint |
| Publicado: |
2024
|
| Materias: | |
| Acceso en línea: | |
| Etiquetas: |
Agregar Etiqueta
Sin Etiquetas, Sea el primero en etiquetar este registro!
|
| _version_ | 1866914709462056960 |
|---|---|
| author | Tan, Chuangchuang Liu, Ping Tao, RenShuai Liu, Huan Zhao, Yao Wu, Baoyuan Wei, Yunchao |
| author_facet | Tan, Chuangchuang Liu, Ping Tao, RenShuai Liu, Huan Zhao, Yao Wu, Baoyuan Wei, Yunchao |
| contents | Recently, the proliferation of increasingly realistic synthetic images generated by various generative adversarial networks has increased the risk of misuse. Consequently, there is a pressing need to develop a generalizable detector for accurately recognizing fake images. The conventional methods rely on generating diverse training sources or large pretrained models. In this work, we show that, on the contrary, the small and training-free filter is sufficient to capture more general artifact representations. Due to its unbias towards both the training and test sources, we define it as Data-Independent Operator (DIO) to achieve appealing improvements on unseen sources. In our framework, handcrafted filters and the randomly-initialized convolutional layer can be used as the training-free artifact representations extractor with excellent results. With the data-independent operator of a popular classifier, such as Resnet50, one could already reach a new state-of-the-art without bells and whistles. We evaluate the effectiveness of the DIO on 33 generation models, even DALLE and Midjourney. Our detector achieves a remarkable improvement of $13.3\%$, establishing a new state-of-the-art performance. The DIO and its extension can serve as strong baselines for future methods. The code is available at \url{https://github.com/chuangchuangtan/Data-Independent-Operator}. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2403_06803 |
| institution | arXiv |
| publishDate | 2024 |
| record_format | arxiv |
| spellingShingle | Data-Independent Operator: A Training-Free Artifact Representation Extractor for Generalizable Deepfake Detection Tan, Chuangchuang Liu, Ping Tao, RenShuai Liu, Huan Zhao, Yao Wu, Baoyuan Wei, Yunchao Computer Vision and Pattern Recognition Recently, the proliferation of increasingly realistic synthetic images generated by various generative adversarial networks has increased the risk of misuse. Consequently, there is a pressing need to develop a generalizable detector for accurately recognizing fake images. The conventional methods rely on generating diverse training sources or large pretrained models. In this work, we show that, on the contrary, the small and training-free filter is sufficient to capture more general artifact representations. Due to its unbias towards both the training and test sources, we define it as Data-Independent Operator (DIO) to achieve appealing improvements on unseen sources. In our framework, handcrafted filters and the randomly-initialized convolutional layer can be used as the training-free artifact representations extractor with excellent results. With the data-independent operator of a popular classifier, such as Resnet50, one could already reach a new state-of-the-art without bells and whistles. We evaluate the effectiveness of the DIO on 33 generation models, even DALLE and Midjourney. Our detector achieves a remarkable improvement of $13.3\%$, establishing a new state-of-the-art performance. The DIO and its extension can serve as strong baselines for future methods. The code is available at \url{https://github.com/chuangchuangtan/Data-Independent-Operator}. |
| title | Data-Independent Operator: A Training-Free Artifact Representation Extractor for Generalizable Deepfake Detection |
| topic | Computer Vision and Pattern Recognition |
| url | https://arxiv.org/abs/2403.06803 |