Text Modality Oriented Image Feature Extraction for Detecting Diffusion-based DeepFake

TL;DR

This paper introduces TOFE, a novel text modality-oriented feature extraction method that effectively detects diffusion-based DeepFake images by leveraging both low-level and high-level image features guided by text embeddings.

Contribution

The paper proposes a new text-guided feature extraction approach, TOFE, to improve detection of diffusion-based DeepFakes by capturing comprehensive image features.

Findings

TOFE outperforms traditional methods across ten diffusion types.

Both low-level and high-level features are crucial for DeepFake detection.

Text embeddings effectively guide image feature representation.

Abstract

The widespread use of diffusion methods enables the creation of highly realistic images on demand, thereby posing significant risks to the integrity and safety of online information and highlighting the necessity of DeepFake detection. Our analysis of features extracted by traditional image encoders reveals that both low-level and high-level features offer distinct advantages in identifying DeepFake images produced by various diffusion methods. Inspired by this finding, we aim to develop an effective representation that captures both low-level and high-level features to detect diffusion-based DeepFakes. To address the problem, we propose a text modality-oriented feature extraction method, termed TOFE. Specifically, for a given target image, the representation we discovered is a corresponding text embedding that can guide the generation of the target image with a specific text-to-image model. Experiments conducted across ten diffusion types demonstrate the efficacy of our proposed method.