Fine-Grained DINO Tuning with Dual Supervision for Face Forgery Detection

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Zhang, Tianxiang, Yu, Peipeng, Xia, Zhihua, Dai, Longchen, Zhou, Xiaoyu, Gao, Hui
Format: Preprint
Published: 2025
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866915620191207424
author Zhang, Tianxiang
Yu, Peipeng
Xia, Zhihua
Dai, Longchen
Zhou, Xiaoyu
Gao, Hui
author_facet Zhang, Tianxiang
Yu, Peipeng
Xia, Zhihua
Dai, Longchen
Zhou, Xiaoyu
Gao, Hui
contents The proliferation of sophisticated deepfakes poses significant threats to information integrity. While DINOv2 shows promise for detection, existing fine-tuning approaches treat it as generic binary classification, overlooking distinct artifacts inherent to different deepfake methods. To address this, we propose a DeepFake Fine-Grained Adapter (DFF-Adapter) for DINOv2. Our method incorporates lightweight multi-head LoRA modules into every transformer block, enabling efficient backbone adaptation. DFF-Adapter simultaneously addresses authenticity detection and fine-grained manipulation type classification, where classifying forgery methods enhances artifact sensitivity. We introduce a shared branch propagating fine-grained manipulation cues to the authenticity head. This enables multi-task cooperative optimization, explicitly enhancing authenticity discrimination with manipulation-specific knowledge. Utilizing only 3.5M trainable parameters, our parameter-efficient approach achieves detection accuracy comparable to or even surpassing that of current complex state-of-the-art methods.
format Preprint
id arxiv_https___arxiv_org_abs_2511_12107
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle Fine-Grained DINO Tuning with Dual Supervision for Face Forgery Detection
Zhang, Tianxiang
Yu, Peipeng
Xia, Zhihua
Dai, Longchen
Zhou, Xiaoyu
Gao, Hui
Computer Vision and Pattern Recognition
The proliferation of sophisticated deepfakes poses significant threats to information integrity. While DINOv2 shows promise for detection, existing fine-tuning approaches treat it as generic binary classification, overlooking distinct artifacts inherent to different deepfake methods. To address this, we propose a DeepFake Fine-Grained Adapter (DFF-Adapter) for DINOv2. Our method incorporates lightweight multi-head LoRA modules into every transformer block, enabling efficient backbone adaptation. DFF-Adapter simultaneously addresses authenticity detection and fine-grained manipulation type classification, where classifying forgery methods enhances artifact sensitivity. We introduce a shared branch propagating fine-grained manipulation cues to the authenticity head. This enables multi-task cooperative optimization, explicitly enhancing authenticity discrimination with manipulation-specific knowledge. Utilizing only 3.5M trainable parameters, our parameter-efficient approach achieves detection accuracy comparable to or even surpassing that of current complex state-of-the-art methods.
title Fine-Grained DINO Tuning with Dual Supervision for Face Forgery Detection
topic Computer Vision and Pattern Recognition
url https://arxiv.org/abs/2511.12107