Evolution of Detection Performance throughout the Online Lifespan of Synthetic Images

TL;DR

This paper investigates how the ability of detectors to identify synthetic images changes over their online lifespan, revealing challenges and proposing a retrieval-assisted method to sustain detection performance.

Contribution

It introduces a retrieval-assisted detection approach that maintains detection efficacy over an image’s online lifespan, improving accuracy and AUC metrics.

Findings

Detectors struggle to distinguish synthetic from real images in the wild.

Detection performance declines as the synthetic image ages online.

Retrieval-assisted detection sustains and improves detection performance.

Abstract

Synthetic images disseminated online significantly differ from those used during the training and evaluation of the state-of-the-art detectors. In this work, we analyze the performance of synthetic image detectors as deceptive synthetic images evolve throughout their online lifespan. Our study reveals that, despite advancements in the field, current state-of-the-art detectors struggle to distinguish between synthetic and real images in the wild. Moreover, we show that the time elapsed since the initial online appearance of a synthetic image negatively affects the performance of most detectors. Ultimately, by employing a retrieval-assisted detection approach, we demonstrate the feasibility to maintain initial detection performance throughout the whole online lifespan of an image and enhance the average detection efficacy across several state-of-the-art detectors by 6.7% and 7.8% for balanced accuracy and AUC metrics, respectively.