Recent advancements in deep generative models have facilitated the creation of photo-realistic images across various tasks. However, these generated images often exhibit perceptual artifacts in specific regions, necessitating manual correction. In this study, we present a comprehensive empirical examination of Perceptual Artifacts Localization (PAL) spanning diverse image synthesis endeavors. We introduce a novel dataset comprising 10,168 generated images, each annotated with per-pixel perceptual artifact labels across ten synthesis tasks. A segmentation model, trained on our proposed dataset, effectively localizes artifacts across a range of tasks. Additionally, we illustrate its proficiency in adapting to previously unseen models using minimal training samples. We further propose an innovative zoom-in inpainting pipeline that seamlessly rectifies perceptual artifacts in the generated images. Through our experimental analyses, we elucidate several practical downstream applications, such as automated artifact rectification, non-referential image quality evaluation, and abnormal region detection in images. The dataset and code are released.
@article{arxiv.2310.05590,
title = {Perceptual Artifacts Localization for Image Synthesis Tasks},
author = {Lingzhi Zhang and Zhengjie Xu and Connelly Barnes and Yuqian Zhou and Qing Liu and He Zhang and Sohrab Amirghodsi and Zhe Lin and Eli Shechtman and Jianbo Shi},
journal= {arXiv preprint arXiv:2310.05590},
year = {2023}
}