LERF-TOGO: Language Embedded Radiance Fields for Zero-Shot Task-Oriented Grasping

This is the official implementation for LERF-TOGO.

Installation

0. Install Nerfstudio dependencies

Follow these instructions up to and including "tinycudann" to install dependencies and create an environment

1. Clone this repo

git clone https://github.com/lerftogo/lerftogo.git

2. Install this repo as a python package

Navigate to this folder and run python -m pip install -e .

3. Install dependencies

cd robot_lerf/graspnet_baseline/knn
python setup.py install 
cd ../pointnet2
python setup.py
cd ../graspnetAPI
pip install -e .

Using LERF-TOGO

Have your data in a folder named data
Run python scripts/gen_grasp.py
Enter data folder name to train/load data
Enter object;part query in lerf query text box
Generate grasps

Bibtex

If you find this useful, please cite the paper!

@inproceedings{lerftogo2023,
 title={Language Embedded Radiance Fields for Zero-Shot Task-Oriented Grasping},
 author={Adam Rashid and Satvik Sharma and Chung Min Kim and Justin Kerr and Lawrence Yunliang Chen and Angjoo Kanazawa and Ken Goldberg},
 booktitle={7th Annual Conference on Robot Learning},
 year = {2023},
}