Object Detection using Deep Learning Techniques

Abstract

This Object Detection Project deals with the identifying of the object in an image (Or Video which is just a bunch of Moving images) provided as an input. Object detection has multiple applications such as face detection, vehicle detection, pedestrian counting, self-driving cars, security systems, etc. The application will be using Tensorflow along with Keras developed by Google. The base language used for development and deployment is Python and C++ tested on Linux. The algorithm to be used in our project for Object detection and classification is ??L?v3 using Nvidia’s CUDA. ??L? is an accurate model which divides images into regions and predicts bounding boxes and probabilities for each region.

 Keywords: Interaction under union (IOU), You look only once (??L?), deformable parts models (DPMs), Probability (Pr), convolutional neural network (CNN)