Prompt-Driven Color Accessibility Evaluation in Diffusion-based Image Generation Models

TL;DR

This paper evaluates how well diffusion-based image generation models produce images accessible to color vision deficiencies, introduces a new metric CVDLoss for assessment, and highlights current models' limitations in addressing color accessibility.

Contribution

It introduces CVDLoss, a novel metric for measuring color accessibility in generated images, and systematically evaluates diffusion models' performance on accessibility prompts.

Findings

Diffusion models often fail to produce accessible images when prompted for color accessibility.

CVDLoss effectively measures structural differences related to color accessibility.

Current models show limited responsiveness to accessibility-focused prompts.

Abstract

Generative models are increasingly integrated into creative workflows. While text-to-image generation excels in visual quality and diversity, color accessibility for users with Color Vision Deficiencies (CVD) remains largely unexplored. Our work systematically evaluates color accessibility in images generated by a common pretrained diffusion model, prompted to improve accessibility across diverse categories. We quantify performance using established, off-the-shelf CVD simulation methods and introduce "CVDLoss", a new metric measuring differences in image gradients indicative of structural detail. We validate CVDLoss against a commonly used daltonization method, demonstrating its sensitivity to color accessibility modifications. Applying CVDLoss to model outputs reveals that existing diffusion models struggle to reliably respond to accessibility-focused prompts. Consequently, our study establishes CVDLoss as a valuable evaluation tool for accessibility-aware image generation and post-processing, offering insights into current generative models' limitations in addressing color accessibility.