SMART RETAIL SHELVES USING DEEP LEARNING AND COMPUTER VISION

Abstract

The paper intends to smartly manage the shelves of retail stores, which is a common problem now- a-day. The previous approach used was by using sensors, but our paper aims to achieve a cost-efficient way of managing the Inventory of the stores by using Computer Vision. By using technologies like Deep Learning, Computer Vision and Python, the paper will not only be cost-efficient but also profitable to the retail stores in predicting material backorders in inventory. Inventory can consists of grocery, vegetables, fruits, etc. based on the requirement of the administrator. In this paper, machine learning models are being investigated for the purpose of proposing a hypothetical model for this problem of an unequal class, in which the frequency of backward events is relatively low compared to the nonlinear objects.