This demo uses person_detection_retail_0013 neural network to detect people. The demo also displays the number of people detected on the frame.
By default, it will use images from /images
folder and change the image every 3 seconds. You can also use color camera input (instead of images from the host) using -cam
argument.
To use a different NN model (eg. MobileNet SSD
or pedestrian_detection_adas_0002
), you would have to change the size
(input size of the NN) and nnPath
variables.
python3 -m pip install -r requirements.txt
python3 main.py