A Mask-RCNN based object detection and captioning framework for industrial videos

Abstract

In today's world, automation and automated systems are growing at a tremendous rate. A huge effort is applied for creating systems and programs that eliminate the need for human effort in day-to-day working. The use of robots in factories to assemble machinery [1], to perform various tasks is not new. However, for labor-intensive industries, that create employment for the masses, it is important to generate reports, find the material availability, perform productivity analysis, etc. Moreover, in any business, it is important to optimize the working of its systems for maximum profit based on minimum investment. To achieve this, productivity analysis is an essential step in any business model.