Research map / Image generation / Fast diffusion sampling
Boundary Guided Learning-Free Semantic Control with Diffusion Models
Ye Zhu, Yu Wu, Zhiwei Deng, Olga Russakovsky, Yan Yan
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 |
|---|---|---|---|
| Use BoundaryDiffusion instead of Asyrp | CelebA-HQ-256 semantic editing · CelebA-HQ-256 | FID: 68.38 vs 63.14 | Improved |
| Use BoundaryDiffusion instead of Asyrp | CelebA-HQ-256 semantic editing · CelebA-HQ-256 | ID: 0.70 vs 0.73 | Improved |
| Use BoundaryDiffusion instead of DiffusionCLIP | CelebA-HQ-256 semantic editing · CelebA-HQ-256 | SC: 93.7% vs 90.4% | Hurt |
| Use BoundaryDiffusion instead of Asyrp | CelebA-HQ-256 semantic editing · CelebA-HQ-256 | S_dir: 0.19 vs 0.17 | Hurt |
| Use BoundaryDiffusion instead of Asyrp | CelebA-HQ-256 semantic editing · CelebA-HQ-256 | User Quality: 41.3% vs 55.5% | Improved |
| Use BoundaryDiffusion instead of Asyrp | CelebA-HQ-256 semantic editing · CelebA-HQ-256 | User Attribute: 44.9% vs 46.9% | Improved |
| Apply one-step boundary guidance in epsilon-space rather than h-space (ablation) | BoundaryDiffusion · CelebA-HQ smile editing | Editing effectiveness | Improved |
| Apply boundary guidance in both epsilon-space and h-space rather than one latent level (ablation) | BoundaryDiffusion · CelebA-HQ smile editing | Editing strength | Improved |
| Add stochastic denoising to mitigate deterministic inversion's distance effect (ablation) | BoundaryDiffusion · Inverted latent editing | Denoised image quality | Improved |
| Edit an inverted latent using a deterministic trajectory rather than a stochastic trajectory (ablation) | Diffusion editing · Inverted latent editing | Denoised image quality | Hurt |
| Use a frozen base model for unseen-domain editing rather than a learning-based method | BoundaryDiffusion · Unseen-domain editing | Editing capability | Hurt |
Earlier methods this paper compares with
Named baselines from earlier accepted papers in this paper's results tables: better than 1 of 1 on most shared results.
- DiffusionCLIP — DiffusionCLIP: Text-Guided Diffusion Models for Robust Image Manipulation (CVPR 2022) · better on 4 of 6 shared results
Matched by method name in results tables read by AI. Authors' own tables tend to favour the newer method, and a comparison can be missed when a method is renamed.
Closest papers by meaning
- Exploring Low-Dimensional Subspace in Diffusion Models for Controllable Image Editing (NeurIPS 2024 · similarity 0.91)
- Drag Your Noise: Interactive Point-based Editing via Diffusion Semantic Propagation (CVPR 2024 · similarity 0.91)
- Unsupervised Region-Based Image Editing of Denoising Diffusion Models (AAAI 2025 · similarity 0.90)
- NoiseCLR: A Contrastive Learning Approach for Unsupervised Discovery of Interpretable Directions in Diffusion Models (CVPR 2024 · similarity 0.89)
- FreeControl: Efficient, Training-Free Structural Control via One-Step Attention Extraction (NeurIPS 2025 · similarity 0.89)
- Inversion-Free Image Editing with Language-Guided Diffusion Models (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.
