Research map / 3D vision / 3D generation
Gen-3Diffusion: Realistic Image-to-3D Generation Via 2D & 3D Diffusion Synergy
Yuxuan Xue, Xianghui Xie, Riccardo Marin, Gerard Pons‐Moll
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 3D-consistent trajectory refinement to 2D sampling (ablation) | Gen3D_avatar · Avatar ablation | CD (cm): 1.57 -> 1.35 | Improved |
| add 3D-consistent trajectory refinement to 2D sampling (ablation) | Gen3D_avatar · Avatar ablation | PSNR: 21.09 -> 21.49 | Improved |
| add 2D multi-view prior as input to the 3D diffusion model (ablation) | Gen3D_avatar · Avatar ablation | CD (cm): 1.75 -> 1.35 | Improved |
| add 2D multi-view prior as input to the 3D diffusion model (ablation) | Gen3D_avatar · Avatar ablation | PSNR: 20.98 -> 21.49 | Improved |
| add 3D-consistent sampling to pretrained multi-view diffusion (ablation) | MVD · 4-view generation | PSNR: 22.32 -> 24.69 | Improved |
| add 3D-consistent sampling to fine-tuned multi-view diffusion (ablation) | MVD_ft · 4-view generation | PSNR: 24.14 -> 24.69 | Improved |
| fine-tune multi-view diffusion on human data | MVD · 4-view generation | PSNR: 22.32 -> 24.14 | Improved |
| vary input elevation from 0 to 30 degrees | Gen3D_object · GSO | PSNR: 22.30 -> 21.46 | No clear effect |
| use joint 2D and 3D diffusion instead of best 2D novel-view diffusion baseline | Gen3D_object · GSO | FID: 64.72 -> 54.12 | Improved |
| template-free 3D-GS instead of SMPL-based reconstruction | Gen3D_avatar · Sizer | CD (cm): 2.35 -> 1.06 | Improved |
Earlier methods this paper compares with
Named baselines from earlier accepted papers in this paper's results tables: better than 0 of 1 on most shared results.
- SiFU — SIFU: Side-view Conditioned Implicit Function for Real-world Usable Clothed Human Reconstruction (CVPR 2024) · better on 0 of 1 shared result
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
- Human-3Diffusion: Realistic Avatar Creation via Explicit 3D Consistent Diffusion Models (NeurIPS 2024 · similarity 0.95)
- Morphable Diffusion: 3D-Consistent Diffusion for Single-image Avatar Creation (CVPR 2024 · similarity 0.92)
- One-2-3-45++: Fast Single Image to 3D Objects with Consistent Multi-View Generation and 3D Diffusion (CVPR 2024 · similarity 0.91)
- Kiss3DGen: Repurposing Image Diffusion Models for 3D Asset Generation (CVPR 2025 · similarity 0.91)
- HumanNorm: Learning Normal Diffusion Model for High-quality and Realistic 3D Human Generation (CVPR 2024 · similarity 0.90)
- Cycle3D: High-quality and Consistent Image-to-3D Generation via Generation-Reconstruction Cycle (AAAI 2025 · similarity 0.90)
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.
