CORE: Consistent Representation Learning for Face Forgery Detection

TL;DR

This paper introduces CORE, a framework that explicitly regularizes representation consistency across augmentations to improve face forgery detection, outperforming existing methods in various experiments.

Contribution

The paper proposes a novel explicit regularization method for consistent representation learning, enhancing face forgery detection accuracy.

Findings

CORE outperforms state-of-the-art methods in in-dataset tests.

CORE demonstrates strong cross-dataset generalization.

Explicit regularization improves representation consistency.

Abstract

Face manipulation techniques develop rapidly and arouse widespread public concerns. Despite that vanilla convolutional neural networks achieve acceptable performance, they suffer from the overfitting issue. To relieve this issue, there is a trend to introduce some erasing-based augmentations. We find that these methods indeed attempt to implicitly induce more consistent representations for different augmentations via assigning the same label for different augmented images. However, due to the lack of explicit regularization, the consistency between different representations is less satisfactory. Therefore, we constrain the consistency of different representations explicitly and propose a simple yet effective framework, COnsistent REpresentation Learning (CORE). Specifically, we first capture the different representations with different augmentations, then regularize the cosine distance of the representations to enhance the consistency. Extensive experiments (in-dataset and cross-dataset) demonstrate that CORE performs favorably against state-of-the-art face forgery detection methods.