This folder contains example code using OpenCV to obtain camera images and perform object detection on the Edge TPU.

This code works on Linux/macOS/Windows using a webcam, Raspberry Pi with the Pi Camera, and on the Coral Dev Board using the Coral Camera or a webcam. For all settings other than the Coral Dev Board, you also need a Coral USB/PCIe/M.2 Accelerator.

1. First, be sure you have completed the setup instructions for your Coral device. If it's been a while, repeat to be sure you have the latest software.

   Importantly, you should have the latest TensorFlow Lite runtime installed (as per the Python quickstart). You can check which version is installed using the pip3 show tflite_runtime command.

2. Clone this Git repo onto your computer or Dev Board:

   mkdir google-coral && cd google-coral
   
   git clone https://github.com/google-coral/examples-camera --depth 1
   

3. Download the models:

   cd examples-camera
   
   sh download_models.sh
   

4. Install the OpenCV libraries:

   cd opencv
   
   bash install_requirements.sh
   

python3 detect.py

By default, this uses the mobilenet_ssd_v2_coco_quant_postprocess_edgetpu.tflite model.

You can change the model and the labels file using flags --model and --labels.