/YOLO3-4-Py

A Python wrapper on Darknet. Compatible with YOLO V3.

Primary LanguagePythonApache License 2.0Apache-2.0

YOLO3-4-Py

PyPI Badge PyPI Badge

A Python wrapper on Darknet. Compatible with latest YOLO V3.

OutputImage Image source: http://absfreepic.com/free-photos/download/crowded-cars-on-street-4032x2272_48736.html

Pre-requisites

  1. Python 3.5
  2. Numpy, cython and pkgconfig pip3 install numpy cython pkgconfig
  3. Optionally, OpenCV 3.x with Python bindings. (Tested on OpenCV 3.4.0)
    • You can use this script to automate Open CV 3.4 installation (Tested on Ubuntu 16.04).
    • It is possible to compile YOLO3-4-Py without OpenCV. (Performance of this approach is less.)
NOTE: OpenCV 3.4.1 has a bug which causes Darknet to fail. Therefore this wrapper would not work with OpenCV 3.4.1.
More details are available at https://github.com/pjreddie/darknet/issues/502

NOTE: It is possible to build yolo3-4-py without OpenCV by setting the environment variable OPENCV=0.

How to run using docker?

  1. Navigate to docker directory.
  2. Copy sample images into the input directory. Or else run input/download_sample_images.sh
  3. Run sh run.sh
  4. Observe the outputs generated in output directory.

How to run in local machine?

  1. Download darknet and compile with OpenCV enabled.
    • Open Makefile of darknet and set OPENCV=1. Then run make.
  2. Set environment variable DARKNET_HOME to download location of darknet.
  3. Add DARKNET_HOME to LD_LIBRARY_PATH. export LD_LIBRARY_PATH=$LD_LIBRARY_PATH:$DARKNET_HOME
  4. Run python3 setup.py build_ext --inplace
  5. Download "yolov3" model file and config files using sh download_models.sh.
  6. Run python3 webcam_test.py or python3 test.py