Research map / Image generation / Fast diffusion sampling
DreamSampler: Unifying Diffusion Sampling and Score Distillation for Image Manipulation
Jeongsol Kim, Geon Yeong Park, Jong Chul Ye
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 noise estimated by the diffusion model in the previous reverse step instead of fresh Gaussian noise (DreamSampler vs DDS) | DDS-style editing · real image editing | fidelity / bias preservation | Improved |
| use DreamSampler for editing instead of baselines | Stable Diffusion v1.5 · Cat->Dog / Horse->Zebra | structure distance: 0.040 -> 0.029 (Cat->Dog, vs EBCA) | Improved |
| DreamSampler on Cat->Cat w/ glasses editing | Stable Diffusion v1.5 · Cat->Cat w/ glasses | CLIP-Acc (%): 81.1% (EBCA) -> 48.3% | Hurt |
| DreamSampler zero-shot inpainting versus TReg | text-guided inpainting · FFHQ 1k glasses | PSNR: 21.97 -> 24.61 | Improved |
| DreamSampler versus TReg | text-guided inpainting · FFHQ 1k glasses | FID: 61.04 -> 27.10 | Improved |
| DreamSampler versus fine-tuned Stable-Inpaint on smile | text-guided inpainting · FFHQ 1k smile | FID: 19.22 -> 24.33 | Hurt |
| single-pass SVG reconstruction with DreamSampler instead of multi-stage PSLD + LIVE + VectorFusion/PDS | SVG restoration from blurry measurement · SVG super-resolution and Gaussian deblur | qualitative fidelity | Improved |
| separate text guidance from data consistency term | text-guided inpainting · FFHQ inpainting | multi-mask text-region control | Improved |
Closest papers by meaning
- DreamCatalyst: Fast and High-Quality 3D Editing via Controlling Editability and Identity Preservation (ICLR 2025 · similarity 0.90)
- Rethinking Score Distillation as a Bridge Between Image Distributions (NeurIPS 2024 · similarity 0.89)
- DREAM: Diffusion Rectification and Estimation-Adaptive Models (CVPR 2024 · similarity 0.88)
- Score Distillation via Reparametrized DDIM (NeurIPS 2024 · similarity 0.88)
- Posterior Distillation Sampling (CVPR 2024 · similarity 0.88)
- A Simple Early Exiting Framework for Accelerated Sampling in Diffusion Models (ICML 2024 · similarity 0.87)
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.
