Project link: https://github.com/thtrieu/darkflow

Object detection is an interesting problem in Computer Vision. There are lots of methods to solve this problem. YOLO, by Joseph Redmon and others, is one of the state-of-the-art techniques. YOLO provides real-time object detection using deep neural networks. Version 2 of YOLO can recognize up to 9,000 different objects with high accuracy in real time. The original YOLO project is programmed in the darknet framework.

In TensorFlow, there is a great implementation of YOLO, called darkflow. The darkflow repository even has the utility that can allow you to export the model and serve on mobile devices.