Leveraging LLMs for On-the-Fly Instruction Guided Image Editing
This addresses the challenge of on-the-fly image editing for users needing quick, instruction-based modifications without prior model preparation, though it is incremental as it builds on existing techniques like image captioning and DDIM inversion.
The paper tackles the problem of editing images based solely on natural language instructions without any preliminary training or fine-tuning, achieving competitive performance by outperforming recent state-of-the-art models on the MAGICBRUSH dataset.
The combination of language processing and image processing keeps attracting increased interest given recent impressive advances that leverage the combined strengths of both domains of research. Among these advances, the task of editing an image on the basis solely of a natural language instruction stands out as a most challenging endeavour. While recent approaches for this task resort, in one way or other, to some form of preliminary preparation, training or fine-tuning, this paper explores a novel approach: We propose a preparation-free method that permits instruction-guided image editing on the fly. This approach is organized along three steps properly orchestrated that resort to image captioning and DDIM inversion, followed by obtaining the edit direction embedding, followed by image editing proper. While dispensing with preliminary preparation, our approach demonstrates to be effective and competitive, outperforming recent, state of the art models for this task when evaluated on the MAGICBRUSH dataset.