Object Detection for Collision Avoidance Demonstration Version
:: Please feel free to utilize the code for your projects with proper citations under the fair use and citation agreement. :: This project is for demonstration purposes, if you'd like to use it commercially, I've built a much more refined and advanced version that is avaliable for download. Please email me directly at haseebk73@gmail.com.
Some Basic Instructions:
-
You will need to download the Yolo.h5 file, place it in the models_data directory, and the yolo.weights file and place it in the yolo_data directory. The weights file exceeds github's space requirements, the weights are therefore not included with this package.
-
Please make sure you have all the dependencies installed for the module.
This is a Driving Obstacle Detection Model, designed to operate with a 'Within Visual Spectrum' camera feed ( 450 -700 nm range ). Can be adapted to multiple frame rates, however +30-FPS is recommended. Useful for detecting other moving and static vehicles, pedestrians, signboards, pets and animals, and other misc road obstacles. Utilized deep adapted C-NN, written using C++ and Python.