TextCraftor: Your Text Encoder Can be Image Quality Controller

arXiv:2403.18978v1 Announce Type: cross
Abstract: Diffusion-based text-to-image generative models, e.g., Stable Diffusion, have revolutionized the field of content generation, enabling significant advancements in areas like image editing and video synthesis. Despite their formidable capabilities, these models are not without their limitations. It is still challenging to synthesize an image that aligns well with the input text, and multiple runs with carefully crafted prompts are required to achieve satisfactory results. To mitigate these limitations, numerous studies have endeavored to fine-tune the …

abstract arxiv capabilities content generation cs.ai cs.cv cs.lg diffusion editing enabling encoder generative generative models image limitations quality stable diffusion synthesis text text-to-image type video