abstract

Object classification is one of the core problem in computer vision. In real world the objects share a relationship with the surrounding. In this project, we attempt to use deep learning techniques to evaluate if a deep learning model can learn the relationship between the object and scene. We attempt to make a deep learning model learn to infer the object from the surrounding. We perform few experiments all of which try to capture this essence some way or the other. Code is made available here https://github.com/thunderock/BlindNet.