search

facebookresearch / segment-anything

The repository provides code for running inference with the SegmentAnything Model (SAM), links for downloading the trained model checkpoints, and example notebooks that show how to use the model.
Apache License 2.0
46.97k stars 5.56k forks source link

3d segmentation in point clouds #342

Open sergiog2 opened 1 year ago

sergiog2 commented 1 year ago

hi, i want to use it for 3d segmentation in point clouds, the model work correctly in ortophotos (Photos in RGB of a point cloud). I ave seen on your web site somthing about 3d reconstruction, i don`t know if you are developing it or there is already a usefull solution. As my goal is to use it to create 3d segments directly from a 3d picture or just recreating it.

akharroubi commented 1 year ago

Here are some implementations of SAM in 3D:

  1. Segment Anything 3D

  2. Segment-lidar

References : 

1: @misc{yang2023sam3d,
title={SAM3D: Segment Anything in 3D Scenes}, 
author={Yunhan Yang, Xiaoyang Wu, Tong He, Hengshuang Zhao and Xihui Liu},
year={2023},
eprint={2306.03908},
archivePrefix={arXiv},
primaryClass={cs.CV}
}

2: @misc{yarroudh:2023:samlidar,
author = {Yarroudh, Anass},
title = {LiDAR Automatic Unsurpervised Segmentation using Segment-Anything Model (SAM) from Meta AI},
year = {2023},
howpublished = {GitHub Repository},
url = {https://github.com/Yarroudh/segment-lidar}
}