nv-tlabs/Difix3D
[CVPR 2025 Oral & Best Paper Finalist] Difix3D+: Improving 3D Reconstructions with Single-Step Diffusion Models
Implements single-step diffusion for real-time 3D artifact removal by distilling diffusion into a one-step inference model, integrating with nerfstudio and gsplat for progressive NeRF/Gaussian splatting refinement. Supports optional reference-image guidance and post-rendering enhancement to sharpen reconstruction artifacts, trained with LPIPS, L2, and gram-based losses via the diffusers framework.
1,113 stars. No commits in the last 6 months.
Stars
1,113
Forks
92
Language
Python
License
—
Category
Last pushed
Jun 28, 2025
Commits (30d)
0
Get this data via API
curl "https://pt-edge.onrender.com/api/v1/quality/diffusion/nv-tlabs/Difix3D"
Open to everyone — 100 requests/day, no key needed. Get a free key for 1,000/day.
Higher-rated alternatives
jayin92/Skyfall-GS
Skyfall-GS: Synthesizing Immersive 3D Urban Scenes from Satellite Imagery
ActiveVisionLab/gaussctrl
[ECCV 2024] GaussCtrl: Multi-View Consistent Text-Driven 3D Gaussian Splatting Editing
Tencent-Hunyuan/Hunyuan3D-2
High-Resolution 3D Assets Generation with Large Scale Hunyuan3D Diffusion Models.
deepseek-ai/DreamCraft3D
[ICLR 2024] Official implementation of DreamCraft3D: Hierarchical 3D Generation with...
caiyuanhao1998/Open-DiffusionGS
Baking Gaussian Splatting into Diffusion Denoiser for Fast and Scalable Single-stage Image-to-3D...