DATA: Multi-Disentanglement based Contrastive Learning for Open-World Semi-Supervised Deepfake Attribution

TL;DR

This paper introduces DATA, a novel contrastive learning framework that improves deepfake attribution by disentangling method-specific features and enhancing generalization to new classes in open-world semi-supervised scenarios.

Contribution

The paper proposes the concept of 'Orthonormal Deepfake Basis' and a multi-disentanglement contrastive learning approach to better distinguish forgery features and handle novel classes.

Findings

Achieves state-of-the-art performance on OSS-DFA benchmark.

Improves accuracy by 2.55% and 5.7% over existing methods.

Effectively disentangles method-specific features from common forgery cues.

Abstract

Deepfake attribution (DFA) aims to perform multiclassification on different facial manipulation techniques, thereby mitigating the detrimental effects of forgery content on the social order and personal reputations. However, previous methods focus only on method-specific clues, which easily lead to overfitting, while overlooking the crucial role of common forgery features. Additionally, they struggle to distinguish between uncertain novel classes in more practical open-world scenarios. To address these issues, in this paper we propose an innovative multi-DisentAnglement based conTrastive leArning framework, DATA, to enhance the generalization ability on novel classes for the open-world semi-supervised deepfake attribution (OSS-DFA) task. Specifically, since all generation techniques can be abstracted into a similar architecture, DATA defines the concept of 'Orthonormal Deepfake Basis' for the first time and utilizes it to disentangle method-specific features, thereby reducing the overfitting on forgery-irrelevant information. Furthermore, an augmented-memory mechanism is designed to assist in novel class discovery and contrastive learning, which aims to obtain clear class boundaries for the novel classes through instance-level disentanglements. Additionally, to enhance the standardization and discrimination of features, DATA uses bases contrastive loss and center contrastive loss as auxiliaries for the aforementioned modules. Extensive experimental evaluations show that DATA achieves state-of-the-art performance on the OSS-DFA benchmark, e.g., there are notable accuracy improvements in 2.55% / 5.7% under different settings, compared with the existing methods.