webcamDetector

==============

Use a Webcam with OpenCV and TensorFlow to detect objects

To use:

1. Upgrade Python

$ brew install --upgrade python3
$ brew link python3
$ python -V

Outputs: 'Python 3.6.2'

2. Install TensorFlow from https://www.tensorflow.org/install/

Test TensorFlow in python:

>>> import tensorflow as tf
>>> tf.__version__

Outputs: '1.3.0'

3. Install TensorFlow Object Detection: https://github.com/tensorflow/models/blob/master/research/object_detection/g3doc/installation.md

Remember to run this from tensorflow/models/object_detection:

export PYTHONPATH=$PYTHONPATH:`pwd`:`pwd`/slim

4. Intall OpenCV

The following should work on MacOS (from https://www.pyimagesearch.com/2016/12/19/install-opencv-3-on-macos-with-homebrew-the-easy-way/):

$ brew tap homebrew/science
$ brew update
$ brew install data_utils
$ brew install ffmpeg
$ brew install opencv3 --with-python3 --with-ffmpeg -v

Test in python:

>>> import cv2
>>> cv2.__version__

Outputs: '3.3.0'

5. download and run:

$ git clone https://github.com/alkari/webcamDetector.git
$ cd webcamDetector
$ python3 webcamDetector.py

The script will download and unpack pretrained model used for detection. It might take a little while depending on your network speed.

5. Press 'Q' on your keyboard to exit.

Recorded detection videos are stored in the "output" directory.