Object Detection and Automatic Image Captioning Using Tensorflow

Abstract

The recent advancement in the field of object detection and image captioning has influenced us to develop this venture. The process of determining objects present in the image is called as object detection. Multiple objects can be detected in a given image. We as humans can easily spot the objects present in the image. The machine needs to understand the specifications of image. Generating the descriptions of images has become popular. It is a difficult task because the image needs to be understood properly and the visual knowledge needs to be converted into descriptive sentence. It is important to detect the objects in the image before captioning it. Object detection falls under one of the international popular research fields. Large scale image classification is possible by building very deep convolutional neural network (CNNs). Pre-trained model is used to predict category of image. The main motive implies training computer to classify an image.