LLM-guided Instance-level Image Manipulation with Diffusion U-Net Cross-Attention Maps
This addresses the problem of flexible and accurate image manipulation for users of generative models, though it is incremental as it builds on existing diffusion and detection methods.
The paper tackles the challenge of precise instance-level control in text-to-image synthesis by proposing a pipeline that uses LLMs, open-vocabulary detectors, and diffusion U-Net cross-attention maps to manipulate images without fine-tuning or input masks, achieving accurate object manipulation as demonstrated in experiments.
The advancement of text-to-image synthesis has introduced powerful generative models capable of creating realistic images from textual prompts. However, precise control over image attributes remains challenging, especially at the instance level. While existing methods offer some control through fine-tuning or auxiliary information, they often face limitations in flexibility and accuracy. To address these challenges, we propose a pipeline leveraging Large Language Models (LLMs), open-vocabulary detectors, cross-attention maps and intermediate activations of diffusion U-Net for instance-level image manipulation. Our method detects objects mentioned in the prompt and present in the generated image, enabling precise manipulation without extensive training or input masks. By incorporating cross-attention maps, our approach ensures coherence in manipulated images while controlling object positions. Our method enables precise manipulations at the instance level without fine-tuning or auxiliary information such as masks or bounding boxes. Code is available at https://github.com/Palandr123/DiffusionU-NetLLM