ThinkDrag: Semantic Drag-Based Image Editing with Visual Reasoning

Key points

• Drag-based image editing provides an intuitive interface for spatially controlled image manipulation by allowing users to specify handle and target points.1
• ThinkDrag is trained to associate point constraints with meaningful object- and scene-level transformations, and can optionally produce an explicit reasoning trace that interprets the intended edit before image generation.1
• Drag Endpoint Markers, motivated by an analysis of cross-image attention, provide an explicit correspondence signal that grounds each drag in image space.1
• To support this framework, we construct a supervised dataset of semantic drag transformations paired with reasoning traces and introduce DragBench++, a benchmark targeting challenging drag-based editing scenarios with reference edit solutions.1

Sources (2)

• [1]ThinkDrag: Semantic Drag-Based Image Editing with Visual Reasoning
  arXiv (AI, ML, NLP, CV, robotics, multi-agent) · Oct 9, 09:58 AM

  We introduce ThinkDrag, a unified multimodal framework for drag-based image editing with visual reasoning.

  Drag-based image editing provides an intuitive interface for spatially controlled image manipulation by allowing users to specify handle and target points.

• [2]ThinkDrag: Semantic Drag-Based Image Editing with Visual Reasoning
  Hugging Face Daily Papers · Oct 9, 12:00 AM · same content

Extractive summary: sentences quoted from the sources.