Improving Text-to-Image Consistency via Automatic Prompt Optimization

arXiv:2403.17804v1 Announce Type: new
Abstract: Impressive advances in text-to-image (T2I) generative models have yielded a plethora of high performing models which are able to generate aesthetically appealing, photorealistic images. Despite the progress, these models still struggle to produce images that are consistent with the input prompt, oftentimes failing to capture object quantities, relations and attributes properly. Existing solutions to improve prompt-image consistency suffer from the following challenges: (1) they oftentimes require model fine-tuning, (2) they only focus on nearby prompt …

abstract advances arxiv consistent cs.cl cs.cv generate generative generative models image images improving object optimization photorealistic photorealistic images progress prompt relations struggle text text-to-image type via