Problem
Research questions and friction points this paper is trying to address.
drag-based editing
spatial control
drag accuracy
semantic consistency
image editing
Drag as Evidence: Motion-Grounded Latent Recomposition for Drag-Based Editing
🤖 AI Summary
Existing drag-based editing methods struggle to balance manipulation precision with generation naturalness. This work proposes MoRe-Drag, which introduces a novel motion-as-evidence latent recomposition mechanism. Specifically, it injects pixel-space warping as a motion prior into the diffusion model sampling trajectory and integrates region-aware latent recomposition with stage-adaptive conditioning to achieve precise editing. Furthermore, the framework incorporates multimodal large language model adaptation to support instruction-free interaction. Extensive evaluations on the DragBench benchmark demonstrate that MoRe-Drag significantly outperforms existing state-of-the-art methods, effectively enhancing dragging accuracy while preserving semantic consistency and visual realism.
📝 Abstract
Modern image editors excel at semantic manipulation and visual synthesis, yet remain limited in precise spatial control, motivating the development of drag-based editing. However, existing drag-based methods often struggle to balance drag accuracy with natural, plausible, and intent-aligned generation. We propose MoRe-Drag, a motion-grounded drag-based editing method. Our key insight is to treat pixel-space warping as coarse motion evidence, and to inject this evidence into the generative sampling trajectory. Specifically, MoRe-Drag performs region-aware latent recomposition over refinement, inpainting, and anchor regions, coupled with stage-adaptive conditioning that progressively shifts from motion-grounded structure formation to semantic refinement. We further support an instruction-free interface by adapting the MLLM-based text encoder for drag-aware instruction inference. Experiments on DragBench-SR and DragBench-DR show that MoRe-Drag substantially improves drag precision over strong base editors and achieves superior drag accuracy among SOTA drag-based methods, while delivering strong semantic consistency and visually realistic results. Code and dataset will be publicly released.
Innovation
Methods, ideas, or system contributions that make the work stand out.
Drag-based Editing
Latent Recomposition
Motion-grounded Generation
Stage-adaptive Conditioning
Instruction-free Interface
🔎 Similar Papers
No similar papers found.
