Global-Local Image Perceptual Score (GLIPS): Evaluating Photorealistic Quality of AI-Generated Images

TL;DR

GLIPS is a new transformer-based image quality metric that aligns closely with human perception, outperforming traditional metrics like FID and SSIM in evaluating AI-generated photorealistic images.

Contribution

This paper introduces GLIPS, a novel perceptual score combining local and global assessments using attention mechanisms and MMD, with an improved scaling method for better interpretability.

Findings

GLIPS correlates more strongly with human judgments than existing metrics.

GLIPS outperforms FID, SSIM, and MS-SSIM in various tests.

The Interpolative Binning Scale improves score interpretability.

Abstract

This paper introduces the Global-Local Image Perceptual Score (GLIPS), an image metric designed to assess the photorealistic image quality of AI-generated images with a high degree of alignment to human visual perception. Traditional metrics such as FID and KID scores do not align closely with human evaluations. The proposed metric incorporates advanced transformer-based attention mechanisms to assess local similarity and Maximum Mean Discrepancy (MMD) to evaluate global distributional similarity. To evaluate the performance of GLIPS, we conducted a human study on photorealistic image quality. Comprehensive tests across various generative models demonstrate that GLIPS consistently outperforms existing metrics like FID, SSIM, and MS-SSIM in terms of correlation with human scores. Additionally, we introduce the Interpolative Binning Scale (IBS), a refined scaling method that enhances the interpretability of metric scores by aligning them more closely with human evaluative standards. The proposed metric and scaling approach not only provides more reliable assessments of AI-generated images but also suggest pathways for future enhancements in image generation technologies.