Jumpat/SegmentAnythingin3D
Segment Anything in 3D with NeRFs (NeurIPS 2023 & IJCV 2025)
Combines Segment Anything Model (SAM) with density-guided inverse rendering to propagate 2D segmentation masks across NeRF views iteratively, enabling complete 3D object extraction from single-view prompts (point or text). Supports multiple NeRF backends including nerfstudio and 3D Gaussian Splatting, with a Dash-based GUI for interactive prompt selection and real-time mask refinement across camera poses.
1,014 stars. No commits in the last 6 months.
Stars
1,014
Forks
65
Language
Python
License
Apache-2.0
Category
Last pushed
May 19, 2025
Commits (30d)
0
Get this data via API
curl "https://pt-edge.onrender.com/api/v1/quality/computer-vision/Jumpat/SegmentAnythingin3D"
Open to everyone — 100 requests/day, no key needed. Get a free key for 1,000/day.
Related tools
weiyithu/NerfingMVS
[ICCV 2021 Oral] NerfingMVS: Guided Optimization of Neural Radiance Fields for Indoor Multi-view Stereo
POSTECH-CVLab/SCNeRF
[ICCV21] Self-Calibrating Neural Radiance Fields
half-potato/nmf
Our method takes as input a collection of images (100 in our experiments) with known cameras,...
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