depthai-gui 2.0.1

DepthAI Pipeline Visualizer GUI

Gen2 Visual Pipeline Editor

This experiment has two components:

1. A visual graph editor that lets you compose a DepthAI pipeline visually and save it as a JSON file.
2. A parser for those JSON files that will turn it into a usable DepthAI pipeline

Editor

The graph editor is a modified version of NodeGraphQt by Johnny Chan.

Install Dependencies:

python3 install_requirements.py

Usage

python3 pipeline_editor.py - Runs the visual pipeline editor

Navigation

• Press Tab to create new nodes
• Right Click To save/load pipeline graphs
• You can find a full list of controls in the NodeGraphQt Documentation

Parser

DAIPipelineGraph is the graph parser.

Demo

python3 demo.py - Runs the included ExampleGraph.json pipeline

Usage

from DAIPipelineGraph import DAIPipelineGraph

pipeline_graph = DAIPipelineGraph( path=pipeline_path )

with dai.Device( pipeline_graph.pipeline ) as device:
  ...

Accessing Pipeline Data

• DAIPipelineGraph.pipeline: A reference to the DepthAI pipeline
• DAIPipelineGraph.nodes: A table of all the nodes. You can access them via the name you put into the "Node Name" field in the editor. Ex: pipeline_graph.nodes["rgb_cam"].setPreviewSize(300,300)
• DAIPipelineGraph.xout_streams: A list of all the names of the XLinkOut streams