Research map / Image generation / Editing and style transfer
Magic Insert: Style-Aware Drag-and-Drop
Nataniel Ruiz, Yuanzhen Li, Neal Wadhwa, Yael Pritch, Michael Rubinstein, David E. Jacobs, Shlomi Fruchter
Read the paperCheck your idea against 137,217 papers
AI-extracted. The measured effects below were read by AI from the paper's full text (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.
Measured effects reported in the paper
| What was changed | Where | Measured | Effect |
|---|---|---|---|
| add ControlNet edge conditioning | Magic Insert · SubjectPlop subject fidelity | overall mean: 0.423 -> 0.495 | Improved |
| add ControlNet edge conditioning | Magic Insert · SubjectPlop subject fidelity | DINO: 0.295 -> 0.514 | Improved |
| add ControlNet edge conditioning | Magic Insert · SubjectPlop style fidelity | overall mean: 0.357 -> 0.381 | Improved |
| add ControlNet edge conditioning | Magic Insert · SubjectPlop | ImageReward: -0.2108 -> -0.1470 | Improved |
| replace InstantStyle ControlNet with Magic Insert ControlNet | InstantStyle ControlNet · SubjectPlop subject fidelity | overall mean: 0.454 -> 0.495 | Improved |
| replace InstantStyle ControlNet with Magic Insert ControlNet | InstantStyle ControlNet · SubjectPlop style fidelity | overall mean: 0.400 -> 0.381 | Hurt |
| replace InstantStyle ControlNet with Magic Insert ControlNet | InstantStyle ControlNet · SubjectPlop | ImageReward: -0.2759 -> -0.1470 | Improved |
| replace StyleAlign ControlNet with Magic Insert ControlNet | StyleAlign ControlNet · SubjectPlop | ImageReward: -0.5180 -> -0.1470 | Improved |
| prefer Magic Insert ControlNet over StyleAlign ControlNet | StyleAlign ControlNet · SubjectPlop user study | user preference: 85% | Improved |
| prefer Magic Insert ControlNet over InstantStyle ControlNet | InstantStyle ControlNet · SubjectPlop user study | user preference: 80% | Improved |
| use edge-conditioned ControlNet instead of VLM prompt | StyleAlign · SubjectPlop subject fidelity | overall mean: 0.383 -> 0.451 | Improved |
| use edge-conditioned ControlNet instead of VLM prompt | InstantStyle · SubjectPlop style fidelity | overall mean: 0.390 -> 0.400 | Improved |
Closest papers by meaning
- MagiCapture: High-Resolution Multi-Concept Portrait Customization (AAAI 2024 · similarity 0.87)
- DragonDiffusion: Enabling Drag-style Manipulation on Diffusion Models (ICLR 2024 · similarity 0.86)
- SigStyle: Signature Style Transfer via Personalized Text-to-Image Models (AAAI 2025 · similarity 0.86)
- DragDiffusion: Harnessing Diffusion Models for Interactive Point-based Image Editing (CVPR 2024 · similarity 0.85)
- Localize, Understand, Collaborate: Semantic-Aware Dragging via Intention Reasoner (NeurIPS 2024 · similarity 0.85)
- StyleGuard: Preventing Text-to-Image-Model-based Style Mimicry Attacks by Style Perturbations (NeurIPS 2025 · similarity 0.85)
Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.
