Viser

  

Viser is a 3D visualization library for computer vision and robotics in Python.

Features include:

• API for visualizing 3D primitives

• GUI building blocks: buttons, checkboxes, text inputs, sliders, etc.

• Scene interaction tools (clicks, selection, transform gizmos)

• Programmatic camera control and rendering

• An entirely web-based client, for easy use over SSH!

The goal is to provide primitives that are (1) easy for simple visualization tasks, but (2) can be composed into more elaborate interfaces. For more about design goals, see the technical report.

Demo reel:

Examples

Install with: pip install viser[examples]

Getting Started

Hello world

The simplest possible viser program: creates a server and adds a red sphere.

Core concepts

Creating a server, adding 3D objects to the scene, and building interactive GUI controls.

Scene Visualization

Coordinate frames

Visualize 3D coordinate systems and hierarchical transformations.

Point cloud visualization

Visualize 3D point clouds with colors.

3D mesh visualization

Load and display 3D meshes from common file formats.

Line visualization

Create line segments and smooth splines for wireframe and path visualization.

Images

Display background images and 3D image textures in the scene.

Batched mesh rendering

Efficiently render many instances of the same mesh with different transforms and colors.

Lighting and shadows

Add directional lights and ambient lighting to illuminate 3D meshes with realistic shadows.

Depth compositing

Display background images that can occlude 3D geometry based on depth values.

Set up direction

Control the global up direction for camera navigation and scene orientation.

Gaussian splats

Viser includes a WebGL-based Gaussian splat renderer.

GUI Controls

Basic GUI controls

Create interactive controls like sliders, buttons, and text inputs.

GUI callbacks

Attach event handlers to GUI elements for real-time interaction.

GUI layouts

Organize GUI controls using folders, tabs, and nested structures for better user experience.

Markdown support

Display rich text content with markdown support.

Modal dialogs

Create popup modal dialogs for user input, confirmations, or detailed information display.

Theming

Customize the appearance with custom titles, logos, and navigation buttons.

Notifications

Display different types of notifications to users including persistent, timed, and loading indicators.

Plotly integration

Embed interactive Plotly charts and graphs directly in the GUI interface.

uPlot integration

Embed lightweight, fast uPlot charts directly in the GUI interface.

Plots as images

Display OpenCV-generated plots as images in the GUI.

User Interaction

Mesh click events

Click on meshes to select them.

Scene pointer events

Capture mouse pointer events to create rays for 3D scene interaction and ray-mesh intersections.

Get renders

Capture rendered images from client viewports for processing or saving.

Camera pose tracking

Monitor and read camera poses from connected clients in real-time.

Programmatic camera control

Control camera position, orientation, and parameters programmatically.

3D GUI elements

Embed GUI controls directly in the 3D scene positioned relative to scene objects.

Demos

Record3D

Parse and stream Record3D captures.

COLMAP

Visualize COLMAP sparse reconstruction outputs.

URDF robot visualizer

Visualize robot models from URDF files with interactive joint controls.

SMPL human model

Visualize SMPL human body models with pose and shape parameter controls.

SMPL skinned mesh

Visualize SMPL human body models using skinned mesh deformation.

Games

Two-player games demonstrating click interactions and game state management.