Contrastive Knowledge Transfer for Deepfake Detection with Limited Data | |
Li, Dongze1,2; Zhuo, Wenqi1,2; Wang, Wei2; Dong, Jing2 | |
2022-11 | |
会议名称 | 26th International Conference on Pattern Recognition (ICPR2022) |
会议日期 | 2022.08.21-2022.08.25 |
会议地点 | Montreal, QC, Canada |
摘要 | Nowadays forensics methods have shown remarkable progress in detecting maliciously crafted fake images. However, without exception, the training process of deepfake detection models requires a large number of facial images. These models are usually unsuitable for real world applications because of their overlarge size and inferiority in speed. Thus, performing dataefficient deepfake detection is of great importance. In this paper, we propose a contrastive distillation method that maximizes the lower bound of mutual information between the teacher and the student to further improve student’s accuracy in a datalimited setting. We observe that models performing deepfake detection, different from other image classification tasks, have shown high robustness when there is a drop in data amount. The proposed knowledge transfer approach is of superior performance compared with vanilla few samples training baseline and other SOTA knowledge transfer methods. We believe we are the first to perform few-sample knowledge distillation on deepfake detection. |
DOI | 10.1109/ICPR56361.2022.9956333 |
收录类别 | EI |
语种 | 英语 |
是否为代表性论文 | 否 |
七大方向——子方向分类 | 图像视频处理与分析 |
国重实验室规划方向分类 | 视觉信息处理 |
是否有论文关联数据集需要存交 | 否 |
引用统计 | |
文献类型 | 会议论文 |
条目标识符 | http://ir.ia.ac.cn/handle/173211/51851 |
专题 | 模式识别实验室 |
通讯作者 | Wang, Wei |
作者单位 | 1.School of Artifcial Intelligence, University of Chinese Academy of Sciences 2.Center for Research on Intelligent Perception and Computing, CASIA |
第一作者单位 | 模式识别国家重点实验室 |
通讯作者单位 | 模式识别国家重点实验室 |
推荐引用方式 GB/T 7714 | Li, Dongze,Zhuo, Wenqi,Wang, Wei,et al. Contrastive Knowledge Transfer for Deepfake Detection with Limited Data[C],2022. |
条目包含的文件 | ||||||
文件名称/大小 | 文献类型 | 版本类型 | 开放类型 | 使用许可 | ||
Contrastive_Knowledg(1186KB) | 会议论文 | 开放获取 | CC BY-NC-SA | 浏览 |
除非特别说明,本系统中所有内容都受版权保护,并保留所有权利。
修改评论