Towards Open-world Generalized Deepfake Detection: General Feature Extraction via Unsupervised Domain Adaptation

TL;DR

This paper introduces a novel training strategy for deepfake detection that enhances generalization in open-world scenarios by leveraging unsupervised domain adaptation techniques to better detect unknown deepfakes.

Contribution

It proposes the OWG-DS training strategy with modules for domain alignment and class boundary separation, advancing deepfake detection in unlabeled, large-scale open-world data.

Findings

Improves detection accuracy across unseen deepfake methods

Enhances model robustness in cross-dataset evaluations

Achieves better domain-invariant feature learning

Abstract

With the development of generative artificial intelligence, new forgery methods are rapidly emerging. Social platforms are flooded with vast amounts of unlabeled synthetic data and authentic data, making it increasingly challenging to distinguish real from fake. Due to the lack of labels, existing supervised detection methods struggle to effectively address the detection of unknown deepfake methods. Moreover, in open world scenarios, the amount of unlabeled data greatly exceeds that of labeled data. Therefore, we define a new deepfake detection generalization task which focuses on how to achieve efficient detection of large amounts of unlabeled data based on limited labeled data to simulate a open world scenario. To solve the above mentioned task, we propose a novel Open-World Deepfake Detection Generalization Enhancement Training Strategy (OWG-DS) to improve the generalization ability of existing methods. Our approach aims to transfer deepfake detection knowledge from a small amount of labeled source domain data to large-scale unlabeled target domain data. Specifically, we introduce the Domain Distance Optimization (DDO) module to align different domain features by optimizing both inter-domain and intra-domain distances. Additionally, the Similarity-based Class Boundary Separation (SCBS) module is used to enhance the aggregation of similar samples to ensure clearer class boundaries, while an adversarial training mechanism is adopted to learn the domain-invariant features. Extensive experiments show that the proposed deepfake detection generalization enhancement training strategy excels in cross-method and cross-dataset scenarios, improving the model's generalization.