Bi-Level Optimization for Self-Supervised AI-Generated Face Detection

TL;DR

This paper presents a self-supervised bi-level optimization approach for face detection that improves generalization to new AI-generated faces by training on photographic images and optimizing pretext tasks.

Contribution

The novel bi-level optimization framework aligns self-supervised learning with AI-generated face detection, enhancing generalization to unseen generative techniques.

Findings

Significantly outperforms existing methods in detection accuracy.

Exhibits strong generalization to unseen face generators.

Effective in both one-class and binary classification tasks.

Abstract

AI-generated face detectors trained via supervised learning typically rely on synthesized images from specific generators, limiting their generalization to emerging generative techniques. To overcome this limitation, we introduce a self-supervised method based on bi-level optimization. In the inner loop, we pretrain a vision encoder only on photographic face images using a set of linearly weighted pretext tasks: classification of categorical exchangeable image file format (EXIF) tags, ranking of ordinal EXIF tags, and detection of artificial face manipulations. The outer loop then optimizes the relative weights of these pretext tasks to enhance the coarse-grained detection of manipulated faces, serving as a proxy task for identifying AI-generated faces. In doing so, it aligns self-supervised learning more closely with the ultimate goal of AI-generated face detection. Once pretrained, the encoder remains fixed, and AI-generated faces are detected either as anomalies under a Gaussian mixture model fitted to photographic face features or by a lightweight two-layer perceptron serving as a binary classifier. Extensive experiments demonstrate that our detectors significantly outperform existing approaches in both one-class and binary classification settings, exhibiting strong generalization to unseen generators.