DIG In: Evaluating Disparities in Image Generations with Indicators for Geographic Diversity

arXiv:2308.06198v3 Announce Type: replace
Abstract: The unprecedented photorealistic results achieved by recent text-to-image generative systems and their increasing use as plug-and-play content creation solutions make it crucial to understand their potential biases. In this work, we introduce three indicators to evaluate the realism, diversity and prompt-generation consistency of text-to-image generative systems when prompted to generate objects from across the world. Our indicators complement qualitative analysis of the broader impact of such systems by enabling automatic and efficient benchmarking of geographic …

abstract arxiv biases cs.cv cs.hc diversity generative image photorealistic prompt results solutions systems text text-to-image type work