#243 · Primary category： Computer Vision
SegmentAnythingin3D
Segment Anything in 3D with NeRFs (NeurIPS 2023 & IJCV 2025)
Project last updated：05/19/25
GitHub Stars
1.0K
Forks
64
Contributors
5
License
Apache-2.0
Why we included this project
SA3D is the official implementation of the NeurIPS 2023 paper that moves Segment Anything's click-to-segment workflow into 3D. Given a trained NeRF, you mark the target object in a single rendered view and the framework propagates the mask through the volume, returning a segmented 3D model in about two minutes with no engineering optimization. A nerfstudio-based branch and a 3D Gaussian Splatting version cut that to seconds, and a 2025 dataset release adds deliberately hard scenes with occlusions, reflective surfaces, and multi-granularity objects to test against. For teams building object-level segmentation into 3D reconstruction pipelines, this is a solid reference to start from.
Articles for this project
No articles for this project yet.
To suggest a topic or contribute an article, contact us.
Related projects in this category
opencv
Open Source Computer Vision Library
RuView
π RuView turns commodity WiFi signals into real-time spatial intelligence, vital sign monitoring, and presence detection — all without a single pixel of video.
PaddleOCR
Turn any PDF or image document into structured data for your AI. A powerful, lightweight OCR toolkit that bridges the gap between images/PDFs and LLMs. Supports 100+ languages.
MinerU
Transforms complex documents like PDFs and Office docs into LLM-ready markdown/JSON for your Agentic workflows.
tesseract
Tesseract Open Source OCR Engine (main repository)