5 papers
Semantic Relation-Enhanced CLIP Adapter for Domain Adaptive Zero-Shot Learning
Jiaao Yu, Mingjie Han, Jinkun Jiang +3
The high cost of data annotation has spurred research on training deep learning models in data-limited scenarios. Existing paradigms, however, fail to balance cross-domain transfer…
Texture, Shape, Order, and Relation Matter: A New Transformer Design for Sequential DeepFake Detection
Yunfei Li, Yuezun Li, Baoyuan Wu +3
Sequential DeepFake detection is an emerging task that predicts the manipulation sequence in order. Existing methods typically formulate it as an image-to-sequence problem, employi…
Forensics Adapter: Unleashing CLIP for Generalizable Face Forgery Detection
Xinjie Cui, Yuezun Li, Delong Zhu +3
We describe Forensics Adapter, an adapter network designed to transform CLIP into an effective and generalizable face forgery detector. Although CLIP is highly versatile, adapting…
Where the Devil Hides: Deepfake Detectors Can No Longer Be Trusted
Shuaiwei Yuan, Junyu Dong, Yuezun Li
With the advancement of AI generative techniques, Deepfake faces have become incredibly realistic and nearly indistinguishable to the human eye. To counter this, Deepfake detectors…
HRGR: Enhancing Image Manipulation Detection via Hierarchical Region-aware Graph Reasoning
Xudong Wang, Jiaran Zhou, Huiyu Zhou +2
Image manipulation detection is to identify the authenticity of each pixel in images. One typical approach to uncover manipulation traces is to model image correlations. The previo…