MASc Seminar: End-to-End Multiview Gesture Recognition for Autonomous Car Parking System

Wednesday, March 27, 2019 11:00 am - 11:00 am EDT (GMT -04:00)

Candidate: Hassene Ben Amara

Title: End-to-End Multiview Gesture Recognition for Autonomous Car Parking System

Date: March 27, 2019

Time: 11:00 AM

Place: EIT 3145

Supervisor(s): Karray, Fakhreddine

Abstract:

The use of hand gestures can be the most intuitive human-machine interaction medium.

The early approaches for hand gesture recognition used device-based methods. These methods use mechanical or optical sensors attached to a glove or markers, which hinders the natural human-machine communication. On the other hand, vision-based methods are not restrictive and allow for a more spontaneous communication without the need of an intermediary between human and machine. Therefore, vision gesture recognition has been a popular area of research for the past thirty years.

Hand gesture recognition finds its application in many areas, particularly the automotive industry where advanced automotive human-machine interface (HMI) designers are using gesture recognition to improve driver and vehicle safety. However, technology advances go beyond active/passive safety and into convenience and comfort. In this context, one of America’s big three automakers has partnered with the Centre of Pattern Analysis and Machine Intelligence (CPAMI) at the University of Waterloo to investigate expanding their product segment through machine learning to provide an increased driver convenience and comfort with the particular application of hand gesture recognition for autonomous car parking.

In this thesis, we leverage the state-of-the-art deep learning and optimization techniques to develop a vision-based multiview dynamic hand gesture recognizer for self-parking system.

We propose a 3DCNN gesture model architecture that we train on a publicly available hand gesture database. We apply transfer learning methods to fine-tune the pre-trained gesture model on a custom made data, which significantly improved the proposed system performance in real world environment. We adapt the architecture of the end-to-end solution to expand the state of the art video classifier from a single image as input (fed by monocular camera) to a multiview 360 feed, offered by a six cameras module. Finally, we optimize the proposed solution to work on a limited resources embedded platform (Nvidia Jetson TX2) that is used by automakers for vehicle based features, without sacrificing the accuracy robustness and real time functionality of the system.