Uncertainty Modeling and its Application in LiDAR-based 3D Object Detection

PhD Thesis Proposal Defence


Title: "Uncertainty Modeling and its Application in LiDAR-based 3D Object
Detection"

by

Mr. Peng YUN


Abstract:

Machine learning has attracted tremendous attention from researchers in various 
fields. In past decades, machine-learning techniques make remarkable progress 
and get great success across a variety of domains, such as robotics, computer 
vision, astronomy, biology, etc. One of the things that makes them so 
fascinating is that they often interact directly with the external world. 
However, the external world is rarely stable. Applying machine-learning 
techniques in critical applications, like autonomous driving, requires not only 
point predictions but also reliable uncertainty measurements. The source of 
uncertainties in machine learning is generally classified into three 
categories: data source, model parameters, and model structures.

In this proposal, we will take LiDAR-based 3D object detection as an instance 
to model the uncertainties in its input point clouds and weight space under the 
context of autonomous driving. For input point clouds, we model the 
uncertainties in extrinsic parameters of a multi-homogeneous-LiDAR system and 
propagate them into each point to improve the robustness of algorithms in 
geometric tasks. In weight space, we evaluate the posterior distribution of 
weight parameters in deep neural networks with Laplacian approximation and 
adopt it as an uncertainty measurement for each parameter. They are further 
used to compute Bayesian constraints for preserving old-task knowledge along 
with knowledge distillation regularizers. Lastly, we will conclude the proposal 
and discuss two remaining problems for further research. It contains modeling 
uncertainties in predictive distributions and limitations of point-estimate 
neural networks. We will discuss two possible solutions: Bayesian Neural 
Networks and non-Bayesian ensemble methods to solve the remaining problems.


Date:			Friday, 5 November 2021

Time:                  	10:00am - 12:00noon

Zoom Meeting: 
https://hkust.zoom.us/j/91725568644?pwd=ZFJRc1BKZXgybmhUUnNxK1AzYmE4QT09

Committee Members:	Dr. Ming Liu (Supervisor)
  			Prof. Dit-Yan Yeung (Chairperson)
 			Dr. Qifeng Chen
 			Prof. Tong Zhang


**** ALL are Welcome ****