OVis: Remote Point Cloud and Human Body Mesh Visualization based on Open3D

OVis is a program that enables remote visualization of point clouds and human body meshes using Open3D. It can be run on a remote server and communicates with the local Open3D visualization program via socketio.

Environment Configuration (Local)

pip install python-socketio
pip install "python-socketio[client]"
pip install numpy
pip install open3d
pip install paramiko
pip instal scipy

Then modify the following fields in config.py:

RemoteIP = '1.2.3.4'
RemoteUser = 'root'
RemotePort = 22
RemotePassword = '123456'
RemotePythonImportPath = '/~/anaconda3/envs/myenv/lib/python3.7'

Environment Configuration (Remote Server)

pip install python-socketio
pip install "python-socketio[client]"
pip install numpy
pip instal scipy

1.Run o3d_server.py on the local machine

python o3d_server.py

This will copy ovis.py to RemotePythonImportPath. Make sure this doesn't overwrite any of your existing files. A new Open3D visualization window will open, which listens for requests on port 5666 and performs the corresponding visualization operations.

2.On the remote/local machine, connect to o3d_server:

from ovis import *

3.Visualize point clouds

opc(np.random.rand(1000, 3))
opc(np.random.rand(1000, 3), 0.2, (1, 0, 0), 'pc2')

4.Visualize SMPL human bodies

osmpl(np.zeros(72), (1, 0, 0), 0.1)

5.Create animated point cloud visualizations

for i in range(10):
    opc('pc1', np.random.rand(1000, 3))
    owait(0.1)

6.Reconnect to o3d_server and clear the visualization

oconnect()
oclear()

7.import ovis locally

To import ovis locally, manually copy ovis.py to your local Python import path and set SERVER_IP in ovis.py to 127.0.0.1.