Grounded Text-to-Image Synthesis with Attention Refocusing
This addresses the issue of poor controllability in text-to-image synthesis for users needing precise image generation, though it is incremental as it builds on existing diffusion models.
The paper tackles the problem of text-to-image synthesis models failing to precisely follow prompts with multiple objects or spatial compositions by proposing attention refocusing losses based on spatial layouts, showing effective improvement in controllability on benchmarks like DrawBench, HRS, and TIFA.
Driven by the scalable diffusion models trained on large-scale datasets, text-to-image synthesis methods have shown compelling results. However, these models still fail to precisely follow the text prompt involving multiple objects, attributes, or spatial compositions. In this paper, we reveal the potential causes in the diffusion model's cross-attention and self-attention layers. We propose two novel losses to refocus attention maps according to a given spatial layout during sampling. Creating the layouts manually requires additional effort and can be tedious. Therefore, we explore using large language models (LLM) to produce these layouts for our method. We conduct extensive experiments on the DrawBench, HRS, and TIFA benchmarks to evaluate our proposed method. We show that our proposed attention refocusing effectively improves the controllability of existing approaches.