Skip to content

Latest commit

 

History

History
63 lines (54 loc) · 1.84 KB

README.md

File metadata and controls

63 lines (54 loc) · 1.84 KB
Raw

Modified code for GraspNet Baseline

Baseline model for "GraspNet-1Billion: A Large-Scale Benchmark for General Object Grasping" (CVPR 2020).

teaser

Requirements

  • Python 3
  • PyTorch 1.6
  • Open3d >=0.8
  • TensorBoard 2.3
  • NumPy
  • SciPy
  • Pillow
  • tqdm

Installation

Get the code.

git clone https://github.com/graspnet/graspnet-baseline.git
cd graspnet-baseline

Install packages via Pip.

pip install -r requirements.txt

Compile and install pointnet2 operators (code adapted from votenet).

cd pointnet2
python setup.py install

Compile and install knn operator (code adapted from pytorch_knn_cuda).

cd knn
python setup.py install

Install graspnetAPI for evaluation.

git clone https://github.com/graspnet/graspnetAPI.git
cd graspnetAPI
pip install .

When using, please use these checkpoints

checkpoint-rs.tar and checkpoint-kn.tar are trained using RealSense data and Kinect data respectively.

Main changes

Camera

A class for camera control was created as RealSenseCameraV2.py

Graspnet

A modified Demo.py was created and named GraspPointsFromRealsense.py. This modified demo.py is a class that outputs grasping poses with an associated grasp score using the RealSenseCameraV2 as input.

Testing

The file Main_using_urx.py contains the prototype for grasping. It uses the urx package to control the ur robot