Research map / 3D vision / 3D generation
DreamCraft3D++: Efficient Hierarchical 3D Generation with Multi-Plane Reconstruction Model
Jingxiang Sun, Cheng Peng, Ruizhi Shao, Yuan-Chen Guo, Xiaochen Zhao, Yangguang Li, Yan‐Pei Cao, Bo Zhang, Yebin Liu
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 learnable top and bottom plane embeddings (ablation) | MP-LRM coarse model · GSO orbiting views | PSNR: 17.119 -> 17.729 | Improved |
| add learnable top and bottom plane embeddings (ablation) | MP-LRM coarse model · GSO orbiting views | LPIPS: 0.174 -> 0.151 | Improved |
| add learnable top and bottom plane embeddings (ablation) | MP-LRM coarse model · GSO orbiting views | Chamfer distance: 0.152 -> 0.131 | Improved |
| add learnable top and bottom plane embeddings (ablation) | MP-LRM coarse model · GSO orbiting views | F-score: 0.754 -> 0.821 | Improved |
| add refinement stage (IP-Adapter VSD) on the coarse mesh | DreamCraft3D++ (Coarse) · GSO orbiting views | PSNR: 17.729 -> 20.251 | Improved |
| add refinement stage on the coarse mesh | DreamCraft3D++ (Coarse) · GSO orbiting views | Chamfer distance: 0.131 -> 0.136 | Hurt |
| add refinement stage on the coarse mesh | DreamCraft3D++ (Coarse) · GSO orbiting views | F-score: 0.821 -> 0.795 | Hurt |
| replace DreamBooth fine-tuning with training-free IP-Adapter | DreamCraft3D texture refinement · texture refinement time | optimization time: 40 minutes -> 10 minutes | Improved |
| replace SDS with VSD for texture refinement (ablation) | texture refinement · qualitative astronaut | visual consistency | Mixed |
| add single-view image-conditioned IP-Adapter to VSD (ablation) | texture refinement · qualitative astronaut | object awareness | Mixed |
| condition IP-Adapter on multi-view images (ablation) | texture refinement · qualitative astronaut | texture consistency | Improved |
| add iterative RGB and normal rendering (ablation) | texture refinement · qualitative astronaut | geometry quality | Improved |
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.
- DreamCraft3D — DreamCraft3D: Hierarchical 3D Generation with Bootstrapped Diffusion Prior (ICLR 2024) · better on 5 of 5 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
- DreamCraft3D: Hierarchical 3D Generation with Bootstrapped Diffusion Prior (ICLR 2024 · similarity 0.91)
- One-2-3-45++: Fast Single Image to 3D Objects with Consistent Multi-View Generation and 3D Diffusion (CVPR 2024 · similarity 0.87)
- DreamControl: Control-Based Text-to-3D Generation with 3D Self-Prior (CVPR 2024 · similarity 0.86)
- 3DTopia-XL: Scaling High-quality 3D Asset Generation via Primitive Diffusion (CVPR 2025 · similarity 0.86)
- Cycle3D: High-quality and Consistent Image-to-3D Generation via Generation-Reconstruction Cycle (AAAI 2025 · similarity 0.86)
- DreamMesh4D: Video-to-4D Generation with Sparse-Controlled Gaussian-Mesh Hybrid Representation (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.
