Review of different techniques for object detection using deep learning

Abstract

Human brain takes less than a minute to identify the location of object inside the image as well as recognize it as soon as it sees to it; but machine needs time and large amount of data to do the same task. Deep neural network based on convolution neural network gives high accuracy and great results in object detection and classification. To train deep neural networks, large amount of data such as (images and videos) and time is required. As computational cost of computer vision is very high, transfer-learning technique, where a model trained on one task is reused on another related task, gives better results. Authors have proposed various deep learning based algorithms for object detection and classification like Region based Convolutional neural network, Fast Region based Convolutional neural network, Faster Region based Convolutional neural network, Mask Region based Convolutional neural network and You Only Look Once. In this paper, a comparative study of different algorithms is given.