UniGenDet: A Unified Generative-Discriminative Framework for Co-Evolutionary Image Generation and Generated Image Detection

TL;DR

UniGenDet introduces a unified framework that enhances both image generation and detection by integrating generative and discriminative models through a novel self-attention mechanism and alignment strategies.

Contribution

It presents the first unified generative-discriminative framework with a symbiotic architecture for co-evolving image creation and detection.

Findings

Achieves state-of-the-art performance on multiple datasets.

Improves interpretability of authenticity identification.

Facilitates higher-fidelity image generation.

Abstract

In recent years, significant progress has been made in both image generation and generated image detection. Despite their rapid, yet largely independent, development, these two fields have evolved distinct architectural paradigms: the former predominantly relies on generative networks, while the latter favors discriminative frameworks. A recent trend in both domains is the use of adversarial information to enhance performance, revealing potential for synergy. However, the significant architectural divergence between them presents considerable challenges. Departing from previous approaches, we propose UniGenDet: a Unified generative-discriminative framework for co-evolutionary image Generation and generated image Detection. To bridge the task gap, we design a symbiotic multimodal self-attention mechanism and a unified fine-tuning algorithm. This synergy allows the generation task to improve the interpretability of authenticity identification, while authenticity criteria guide the creation of higher-fidelity images. Furthermore, we introduce a detector-informed generative alignment mechanism to facilitate seamless information exchange. Extensive experiments on multiple datasets demonstrate that our method achieves state-of-the-art performance. Code: \href{https://github.com/Zhangyr2022/UniGenDet}{https://github.com/Zhangyr2022/UniGenDet}.