中国科学院机构知识库网格
Chinese Academy of Sciences Institutional Repositories Grid
首页 机构 成果 学者
  
  1. CAS IR Grid
  2. 自动化研究所
  3. 中国科学院自动化研究所
  4. 智能感知与计算研究中心
Test-time Forgery Detection with Spatial-Frequency Prompt Learning

文献类型：期刊论文

作者Duan, Junxian1,2,3; Ai, Yuang1,2,3; Liu, Jipeng4; Huang, Shenyuan1; Huang, Huaibo1,2,3; Cao, Jie1,2; He, Ran1,2,3
刊名INTERNATIONAL JOURNAL OF COMPUTER VISION
出版日期2024-08-13
页码16
关键词Face forgery detection Spatial-frequency prompt learning Test-time training Generalization Diffusion model
ISSN号0920-5691
DOI10.1007/s11263-024-02208-2
通讯作者He, Ran(rhe@nlpr.ia.ac.cn)
英文摘要The significance of face forgery detection has grown substantially due to the emergence of facial manipulation technologies. Recent methods have turned to face detection forgery in the spatial-frequency domain, resulting in improved overall performance. Nonetheless, these methods are still not guaranteed to cover various forgery technologies, and the networks trained on public datasets struggle to accurately quantify their uncertainty levels. In this work, we design a Dynamic Dual-spectrum Interaction Network that allows test-time training with uncertainty guidance and spatial-frequency prompt learning. RGB and frequency features are first interacted in multi-level by using a Frequency-guided Attention Module. Then these multi-modal features are merged with a Dynamic Fusion Module. As a bias in the fusion weight of uncertain data during dynamic fusion, we further exploit uncertain perturbation as guidance during the test-time training phase. Furthermore, we propose a spatial-frequency prompt learning method to effectively enhance the generalization of the forgery detection model. Finally, we curate a novel, extensive dataset containing images synthesized by various diffusion and non-diffusion methods. Comprehensive evaluations of experiments show that our method achieves more appealing results for face forgery detection than recent state-of-the-art methods.
资助项目National Natural Science Foundation of China[U21B2045] ; National Natural Science Foundation of China[U20A20223] ; National Natural Science Foundation of China[32341009] ; National Natural Science Foundation of China[62206277] ; Youth Innovation Promotion Association CAS[2022132] ; Beijing Nova Program[20230484276]
WOS研究方向Computer Science
语种英语
WOS记录号WOS:001290129700002
出版者SPRINGER
资助机构National Natural Science Foundation of China ; Youth Innovation Promotion Association CAS ; Beijing Nova Program
源URL[http://ir.ia.ac.cn/handle/173211/59318]  
专题自动化研究所_智能感知与计算研究中心
通讯作者He, Ran
作者单位1.Chinese Acad Sci, Inst Automat, Beijing, Peoples R China
2.CASIA, State Key Lab Multimodal Artificial Intelligence S, Beijing, Peoples R China
3.Univ Chinese Acad Sci, Beijing, Peoples R China
4.Chinese Acad Sci, Inst Informat Engn, Beijing, Peoples R China
推荐引用方式
GB/T 7714		 Duan, Junxian,Ai, Yuang,Liu, Jipeng,et al. Test-time Forgery Detection with Spatial-Frequency Prompt Learning[J]. INTERNATIONAL JOURNAL OF COMPUTER VISION,2024:16.
APA Duan, Junxian.,Ai, Yuang.,Liu, Jipeng.,Huang, Shenyuan.,Huang, Huaibo.,...&He, Ran.(2024).Test-time Forgery Detection with Spatial-Frequency Prompt Learning.INTERNATIONAL JOURNAL OF COMPUTER VISION,16.
MLA Duan, Junxian,et al."Test-time Forgery Detection with Spatial-Frequency Prompt Learning".INTERNATIONAL JOURNAL OF COMPUTER VISION (2024):16.

入库方式： OAI收割

来源：自动化研究所

浏览0
下载0
收藏0
其他版本

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。