A SURVEY OF MULTIPLE OBJECT TRACKING IN AUTONOMOUS DRIVING

PhD Qualifying Examination


Title: "A SURVEY OF MULTIPLE OBJECT TRACKING IN AUTONOMOUS DRIVING"

by

Mr. Sukai WANG


Abstract:

Multiple object tracking (MOT) plays an important role in autonomous driving. 
The goal of the MOT is to analyze a sequence of sensor perception in order to 
identify and track objects belonging to one or more categories. In this survey, 
we will firstly introduce the MOT, research background, current object 
detection development, and common sensors and data types. Then problem 
definition and the commonly used metrics are identified, and the popular 
datasets with the simulation environment are provided. Next, we will introduce 
various top-performing methods used for solving different challenges  in MOT, 
with their comparison and analysis results. Finally, I will present several 
promising future research directions, including multi-sensor fusion, generative 
adversarial networks (GANs), multi-scan tracking in Spatio-temporal map, and a 
big data-driven prior learnable method.


Date:			Thursday, 11 June 2020

Time:                  	10:00am - 12:00noon

Zoom meeting:		https://hkust.zoom.us/j/9127341968

Committee Members:	Dr. Ming Liu (Supervisor)
 			Dr. Qifeng Chen (Chairperson)
 			Dr. Shaojie Shen (ECE)
 			Prof. Tong Zhang


**** ALL are Welcome ****