half-potato/nmf
Our method takes as input a collection of images (100 in our experiments) with known cameras, and outputs the volumetric density and normals, materials (BRDFs), and far-field illumination (environment map) of the scene.
This project helps 3D artists, game developers, or product designers accurately reconstruct detailed 3D scenes from a collection of existing images. You provide various photos of an object or scene, taken from different angles with known camera positions. The project then generates a complete 3D model, including its shape, surface materials, and the lighting conditions of the original environment.
Use this if you need to create realistic 3D models and accurately capture the material properties and lighting of real-world objects or scenes from multiple photographs.
Not ideal if you're looking for a simple, automated 3D scanning solution that doesn't require technical configuration or if your input images lack precise camera information.
Stars
58
Forks
5
Language
Jupyter Notebook
License
MIT
Category
Last pushed
Dec 02, 2025
Commits (30d)
0
Get this data via API
curl "https://pt-edge.onrender.com/api/v1/quality/computer-vision/half-potato/nmf"
Open to everyone — 100 requests/day, no key needed. Get a free key for 1,000/day.
Higher-rated alternatives
weiyithu/NerfingMVS
[ICCV 2021 Oral] NerfingMVS: Guided Optimization of Neural Radiance Fields for Indoor Multi-view Stereo
Jumpat/SegmentAnythingin3D
Segment Anything in 3D with NeRFs (NeurIPS 2023 & IJCV 2025)
POSTECH-CVLab/SCNeRF
[ICCV21] Self-Calibrating Neural Radiance Fields
synthesiaresearch/humanrf
Official code for "HumanRF: High-Fidelity Neural Radiance Fields for Humans in Motion"
HankYe/PAGCP
[T-PAMI'23] PAGCP for the compression of YOLOv5