$ sudo apt update
$ sudo apt upgrade
$ sudo apt install python3-opencv
$ python
>>> import cv2
>>> print(cv2.__version__)
sudo apt install python3-opencv
$ sudo apt-get update
$ sudo apt-get dist-upgrade
$ git clone https://github.com/EdjeElectronics/TensorFlow-Lite-Object-Detection-on-Android-and-Raspberry-Pi.git
$ mv TensorFlow-Lite-Object-Detection-on-Android-and-Raspberry-Pi tflite1
$ cd tflite1
$ sudo pip3 install virtualenv
$ python3 -m venv tflite1-env
$ source tflite1-env/bin/activate
$ bash get_pi_requirements.sh
$ wget https://storage.googleapis.com/download.tensorflow.org/models/tflite/coco_ssd_mobilenet_v1_1.0_quant_2018_06_29.zip
$ unzip coco_ssd_mobilenet_v1_1.0_quant_2018_06_29.zip -d Sample_TFLite_model
$ python TFLite_detection_video.py --modeldir=Sample_TFLite_model --video=test.mp4
$ python TFLite_detection_video.py --modeldir=Sample_TFLite_model --video=dog.mp4
python TFLite_detection_video.py --modeldir=Sample_TFLite_model --video=dog.mp4
TFLite_detection_video.py
The TFLite_detection_webcam.py program worked (with line self.stream = cv2.VideoCapture(0)) after I enabled Interface Options | Legacy camera using raspi-config.
$ sudo raspi-config
TFLite_detection_webcam.py
Logitech QuickCam for Notebooks (from Goodwill)
Performance improvement from 5.17 frames per second to 28.59 frames per second
