Panopti:
_{Interactive 3D Visualization in Python}

[ Documentation ] - [ PyPI ]

Panopti is a Python package for interactive 3D visualization that is designed so that users only need to write Python code, making it painless to setup interactive experiments and scenes. All code examples and demos throughout the documentation are achieved purely using Panopti in Python -- no JavaScript required!

Panopti offers several features:

• ✅ Remote SSH compatible
• ✅ Headless rendering
• ✅ Geometric primitives: meshes, point clouds, animated geometry, arrows, etc.
• ✅ Interactive UI elements with Python callbacks
• ✅ Programmable events: on-click, inspection tool, transformation gizmo, camera update, etc.
• ✅ Convenience features: exporting geometry, embedding Plotly figures
• ✅ Interactive web console that lets you debug with your Python script's state
• ✅ Material customization

See the docs for more!

Install from pip:

pip install panopti

Simple example

First start a Panopti server in a separate terminal:

python -m panopti.run_server --host localhost --port 8080

Then you can easily define your scene, for example:

import panopti
import trimesh # just for io
import numpy as np

# create panopti client that connects to server:
viewer = panopti.connect(server_url="http://localhost:8080", viewer_id='client') 
# open viewer in browser: http://localhost:8080/?viewer_id=client

mesh = trimesh.load('./examples/demosthenes.obj')
verts, faces = mesh.vertices, mesh.faces

# add a mesh to the scene:
viewer.add_mesh(
    vertices=verts,
    faces=faces,
    name="Statue"
)

def callback_button(viewer):
    # Update mesh vertices on button press by adding Gaussian noise
    statue = viewer.get('Statue')
    noise = np.random.normal(scale=0.05, size=statue.vertices.shape)
    new_verts = statue.vertices + noise
    statue.update(vertices=new_verts)

viewer.button(callback=callback_button, name='Click Me!')

viewer.hold() # prevent script from terminating

For more examples see Documentation or /examples

Installation

To install from pip:

pip install panopti

To install from source:

git clone https://github.com/ArmanMaesumi/panopti

# build frontend viewer
cd panopti/frontend 
npm install
npm run build
cd ..

# install python package
pip install .

Dependencies

Core dependencies:

pip install numpy eventlet requests flask flask-socketio python-socketio[client] tomli msgpack trimesh

Optional dependencies:

pip install matplotlib      # for colormap utilities
pip install plotly==5.22.0  # for plotly figure support

Doc-related dependencies:

pip install mkdocs mkdocs-material mkdocstrings mkdocstrings-python

Development

Details for running the local development workflow can be found in: /frontend/README.md

Cite as

@misc{panopti,
  title = {Panopti},
  author = {Arman Maesumi},
  note = {https://github.com/ArmanMaesumi/panopti},
  year = {2025}
}