This example demonstrates how to run 3 stage inference (3-series, 2-parallel) on DepthAI using Gen2 Pipeline Builder.
Peek.2021-01-28.14-45.mp4
Original OpenVINO demo, on which this example was made, is official here from Intel and implemented with the NCS2 with nice diagrams and explanation, here, from @LCTyrell.
Figure: @LCTyrell
Install requirements:
python3 -m pip install -r requirements.txt
usage: main.py [-h] [-nd] [-cam] [-vid VIDEO]
optional arguments:
-h, --help show this help message and exit
-nd, --no-debug Prevent debug output
-cam, --camera Use DepthAI 4K RGB camera for inference (conflicts with -vid)
-vid VIDEO, --video VIDEO
Path to video file to be used for inference (conflicts with -cam)
To use with a video file, run the script with the following arguments
python3 main.py -vid ./demo.mp4
To use with DepthAI 4K RGB camera, use instead
python3 main.py -cam