2D IMAGE SEMANTIC SEGMENTATION FOR SELF-DRIVING CAR USING CONVOLUTION NEURAL NETWORK

Semantic Segmentation is the process of clustering image parts together which belong to the same class. Visual Perception is to interpret the surrounding environment using camera sensors, calibration in Self-Driving. In this paper an effort has been made to focus on detecting objects in 2D images captured via camera on the Self-Driving Car through the concepts of Visual Perception and Semantic Segmentation. The Cityscape data set comprises 2D images which is used as an input and passed though Convolution Neural Network (CNN), a set of convolution layers which make sure there is no data loss and gains accuracy in object detection.