- Download the model file from here
- Create a model directory and save the model file in that directory
- Edit line 45 and 47 in
main.py
file. Add the openvino model and library path based on your system - Run
main.py
file
- Download the model file from here
- Create a model directory and save the model file in that directory
- Create a sample database, having the following directory structure
/data
|-- class_1 - 1.jpg
|
|-- class_2 - 2.jpg
.
.
- Run
create_embeddings.py
file to generate the embeddings for the custom dataset
- Edit line 45 and 47 in
main.py
file. Add the openvino model and library path based on your system - Run
main.py
file.
NOTE - For inference on a video file. Save the video file in the videos directory. The output file for each video will be stored in the output directory.
You can checkout the processed video here -