A project which uses deep learning to detect and estimate the distance of the detected objects from the monocular camera. An alarm is raised the detected object is in a distance range considered as dangerous.
In this poject we used the TensorFlow implementation of Yolov3 architecture from pythonlesson. We trained our model on the Kitti dataset for this model to be detect the road object. This dataset has 9 classes which are 'car', 'Van', 'Truck','person', 'Person_sitting', 'Cyclist', 'Tram', 'Misc' and 'DontCare'