ABSTRACT
In the cutting-edge and swift world, finding a precise and proficient object recognition for headway in computer vision frameworks has been an integral part. With the emerge of faster (due to faster and better hardware too) and increasingly accurate deep learning techniques the prediction for finding a good and accurate model is of paramount importance. In this undertaking, we are aiming to incorporate the best possible technique to tackle object detection, with the goal of achieving higher accuracy in our Multiple Object Detection project. The major motivation in our task is to make colossal contribution to increasing the accuracy of object detection. In this venture, we resort to four models to tackle the problem, mainly Mask-RCNN (Mask Region Convolution Neural Network), Faster-RCNN (Faster Region Convolution Neural Network), SSD (Single Shot MultiBox Detection), and YOLO (You Look Only Once) trained on the most publicly and freely accessible dataset (MS-COCO).
