Analysis of Instance Segmentation using Mask-RCNN

Abstract

Object detection has been one of the greatest achievement in the field of Machine learning. Most of the models in the domain carry out the process of identifying the character or object using a bounding box. In the recent years, detection of the object using Instance segmentation has been in limelight. However, no interest has been showed towards the data present in the field of Entertainment. The focus of this paper is to carry out pixel level comparison and to study the behaviour of the model by analysing the result obtained at various different instances.