Towards Futuristic Visual StoryTelling: Authoring Data-Driven Infographics in Augmented Reality

PhD Thesis Proposal Defence


Title: "Towards Futuristic Visual StoryTelling: Authoring Data-Driven 
Infographics in Augmented Reality"

by

Mr. Zhutian CHEN


Abstract:

Visual data-driven  storytelling  is  concerned  with  effective 
communication about data through visualization. Recent advances in 
Augmented Reality (AR) have shed new light on data-driven storytelling, 
offering exciting possibilities for telling attractive, engaging, 
creative, and immersive stories.  However, creating such kind of AR data 
stories is demanding. Mainstream solutions for creating AR content require 
users to master considerable knowledge of different domains (e.g., data 
visualization, computer graphics, computer vision, and human-machine 
interaction) and skills of various tools (e.g., D3, Unity, ARKit). Past 
work has rarely investigated authoring visual data-driven stories in AR 
environments. As a first step, we explore the approaches to enable 
authoring infographics, a popular format for data-driven storytelling, in 
AR environments.

The first research problem we addressed focuses on 3D infographics. By 
addressing the challenge of striking a balance between the expressivity 
and efficiency, we design and implement MARVisT. MARVisT is a mobile 
authoring tool that leverages information from reality to assist 
non-experts in creating 3D personal visualization in mobile AR. An example 
gallery is presented to demonstrate the expressiveness and a user study 
with non-experts is conducted to evaluate the authoring experience of 
MARVisT. Our second focusing area is extending 2D infographics using AR. 
AR techniques can extend 2D infographics to display more details and 
updated data even the infographics are printed. To extend existing printed 
infographics, we first focus on timeline infographics and propose a deep 
learning-based approach. The approach automatically parses a bitmap 
timeline infographic to generates the new timeline with new data following 
the style of the existing one. Quantitative evaluation of our approach 
over two datasets is reported, and examples are presented to demonstrate 
the performance qualitatively. Finally, we briefly introduce our ongoing 
work on enabling users to create inherent AR extendable 2D infographics 
and discuss future potential research.


Date:			Thursday, 22 August 2019

Time:                  	2:00pm - 4:00pm

Venue:                  Room 3494
                         lifts 25/26

Committee Members:	Prof. Huamin Qu (Supervisor)
 			Prof. Andrew Horner (Chairperson)
 			Dr. Xiaojuan Ma
 			Dr. Sai-Kit Yeung (ISD)


**** ALL are Welcome ****