Research map / 3D vision / 3D generation
Repaint123: Fast and High-quality One Image to 3D Generation with Progressive Controllable Repainting
Junwu Zhang, Zhenyu Tang, Yatian Pang, Xinhua Cheng, Peng Jin, Yida Wei, xing zhou, munan ning, Li Yuan
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 mutual attention to vanilla repainting (ablation) | Repaint123 · Test-alpha | CLIP-Similarity: 0.71 -> 0.78 | Improved |
| add image prompt for classifier-free guidance (ablation) | Repaint123 · Test-alpha | CLIP-Similarity: 0.78 -> 0.84 | Improved |
| add adaptive repainting mask (visibility-based) (ablation) | Repaint123 · Test-alpha | CLIP-Similarity: 0.84 -> 0.88 | Improved |
| add progressive repainting refinement to the coarse 3D Gaussians (ablation) | DreamGaussian coarse stage · Test-alpha | LPIPS: 0.133 -> 0.049 | Improved |
| image prompt versus text prompt or textual inversion (ablation) | Repaint123 · RealFusion15 | CLIP-Similarity: 0.79 -> 0.85 | Improved |
| increase view angle interval from 40 to 60 degrees (ablation) | Repaint123 · Test-alpha | CLIP-Similarity: 0.881 -> 0.888 | Mixed |
| Gaussian Splatting coarse stage versus NeRF-based methods on reference-view reconstruction | Repaint123 · Test-alpha | PSNR: 22.38 vs 24.69 (Magic123) | Hurt |
| use NeRF instead of Gaussian Splatting for the coarse stage (ablation) | Repaint123 · RealFusion15 | PSNR: 19.00 -> 20.27 | Improved |
Earlier methods this paper compares with
Named baselines from earlier accepted papers in this paper's results tables: better than 1 of 2 on most shared results.
- Magic123 — Magic123: One Image to High-Quality 3D Object Generation Using Both 2D and 3D Diffusion Priors (ICLR 2024) · better on 1 of 4 shared results
- DreamGaussian — DreamGaussian: Generative Gaussian Splatting for Efficient 3D Content Creation (ICLR 2024) · better on 2 of 2 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
- Cycle3D: High-quality and Consistent Image-to-3D Generation via Generation-Reconstruction Cycle (AAAI 2025 · similarity 0.89)
- Prune and Repaint: Content-Aware Image Retargeting for any Ratio (NeurIPS 2024 · similarity 0.88)
- Paint3D: Paint Anything 3D with Lighting-Less Texture Diffusion Models (CVPR 2024 · similarity 0.88)
- One-2-3-45++: Fast Single Image to 3D Objects with Consistent Multi-View Generation and 3D Diffusion (CVPR 2024 · similarity 0.86)
- DreamCraft3D: Hierarchical 3D Generation with Bootstrapped Diffusion Prior (ICLR 2024 · similarity 0.86)
- Unique3D: High-Quality and Efficient 3D Mesh Generation from a Single Image (NeurIPS 2024 · similarity 0.86)
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.
