Digital transformation with glassware

This project aims to develop a new deep learning algorithm based on various existing CNN-based algorithms, like SPPNet, Fast RCNN, Faster RCNN, YOLO, RetinaNet, CornerNet, etc. to detect the specific machine, forklift, in an image stream and classify the images by the type of forklift. Meanwhile, the company’s existing cloud environment and the speed requested by a real-time system must be considered. The project will resolve a couple of problems. For example, how to recognize a forklift if the significant component is blocked in natural images, how to identify the type of different forklift with similar features, how to make sure the model computes quickly enough to satisfy the request of a real-time system, and so on.

Faculty Supervisor:

Ioannis Mitliagkas

Student:

Partner:

Groupe ENO

Discipline:

Computer science

Sector:

Education

University:

Université de Montréal

Program: