Research map / Image generation / Editing and style transfer
Blended Diffusion for Text-Driven Editing of Natural Images
Check your idea against 108,967 papers
AI-extracted summary. Written by AI from the paper's title only (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 |
|---|---|---|---|
| Blend the generated foreground latent with the correspondingly noised input background at each diffusion step | Blended Diffusion · User study | Background preservation: 4.73±0.61 vs. Local CLIP GD 3.11±1.24 | Helped |
| Use blended diffusion instead of baseline editing methods | Text-driven local image editing · User study | Realism: 3.93±1.08 vs. 3.50±1.19 | Helped |
| Use blended diffusion instead of baseline editing methods | Text-driven local image editing · User study | Text match: 4.63±0.77 vs. 3.86±1.32 | Helped |
| Add extending augmentations when computing CLIP guidance (ablation) | Blended Diffusion · Text-driven local image editing | Visual plausibility and coherence | Helped |
| Replace DDPM sampling with DDIM sampling (ablation) | Blended Diffusion · 100 diffusion steps | Result quality | Hurt |
| Replace DDPM sampling with DDIM sampling (ablation) | Blended Diffusion · Fewer than 50 diffusion steps | Result quality | Helped |
| Increase the scribble-editing diffusion steps (ablation) | Blended Diffusion scribble editing · Scribble-guided editing | Realism and diversity: k=80; k=20 | Helped |
| Increase the scribble-editing diffusion steps (ablation) | Blended Diffusion scribble editing · Scribble-guided editing | Scribble color preservation: k=80; k=20 | Hurt |
Closest papers by meaning
- An Item Is Worth a Prompt: Versatile Image Editing with Disentangled Control (AAAI 2025 · similarity 0.89)
- Towards Efficient Diffusion-Based Image Editing with Instant Attention Masks (AAAI 2024 · similarity 0.89)
- Progressive Text-to-Image Diffusion with Soft Latent Direction (AAAI 2024 · similarity 0.89)
- Inversion-Free Image Editing with Language-Guided Diffusion Models (CVPR 2024 · similarity 0.89)
- Textualize Visual Prompt for Image Editing via Diffusion Bridge (AAAI 2025 · similarity 0.89)
- De-Diffusion Makes Text a Strong Cross-Modal Interface (CVPR 2024 · similarity 0.89)
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.