UWindsor Together: Student Mental Health and Remote Learning Services

MSc Thesis Proposal Announcement of Hardik Sonetta:"Bridging the Simulation-to-Reality Gap: Adapting Simulation Environment for Object Recognition"

Monday, February 8, 2021 - 10:00 to 12:00

SCHOOL OF COMPUTER SCIENCE 

The School of Computer Science is pleased to present...

MSc Thesis Proposal by: Hardik Sonetta

 
Date: Monday February 8, 2021
Time: 10:00 am to 12:00 pm
Passcode: If interested in attending the event, contact the Graduate Secretary at csgradinfo@uwindsor.ca
 

Abstract: 

Rapid advancements in object recognition have created a huge demand for labeled datasets for the task of training, evaluation, and validation of different techniques. Due to the wide range of applications, object models in the datasets need to cover both variations in geometric features and diverse conditions in which sensory inputs are obtained. The need to manually labeling the models is also cumbersome. As a result, it becomes difficult for researchers to gain access to adequate datasets for the development of new methods or algorithms. In comparison, computer simulation has been considered as a cost-effective solution to generate simulated data for the training and validation of object recognition techniques. However, its effectiveness has been the major concern due to a problem commonly known as the reality gap, which emphasizes the differences that exist between real and computer-synthesized datasets. Aimed at bridging the reality gap, this presentation first examines the influential parameters that cause the problem and then proposes to adjust the setting of simulation to not only replicate the objects but also the environment that matches with the real-world scenario. In addition, it includes a system structure to explain how to retrieve information of the real world and to incorporate this information in the setting of environmental properties in simulation. It is anticipated that the proposed approach is going to enable the rendering of realistic data with ground-truth labels, thus making simulated datasets a cost-effective and efficient alternative. 
 
Keywords: Simulation, Synthetic Dataset, Object Recognition
 

MSc Thesis Committee: 

Internal Reader: Dr. Imran Ahmad
External Reader: Dr. Mohammed Khalid
Advisor: Dr. Dan Wu
Co-Advisor: Dr. Xiaobu Yuan
 

MSc Thesis Proposal Announcement

 
5113 Lambton Tower 401 Sunset Ave. Windsor ON, N9B 3P4 (519) 253-3000 Ext. 3716 csgradinfo@uwindsor.ca