Adapting Computational Creativity Support From Desktop to Virtual Reality

The Hong Kong University of Science and Technology
Department of Computer Science and Engineering


PhD Thesis Defence


Title: "Adapting Computational Creativity Support From Desktop to Virtual 
Reality"

By

Miss Linping YUAN


Abstract:

Virtual reality (VR) holds the promise of transforming various domains through 
immersive experiences. However, the widespread adoption of VR has been 
significantly slowed by the high barriers to content creation, which require 
both technical skills and specialized knowledge in immersive experience design. 
This thesis investigates how computational creativity support can lower these 
barriers and make VR content creation more accessible. Specifically, it 
explores where whether, and how computational approaches developed for desktop 
content can be adapted to address the unique challenges of VR content creation, 
given the scarcity of VR-specific knowledge and data.

The thesis consists of three main parts. The first part involves understanding 
the current practices and challenges faced by VR creators to explore where 
creativity support is needed. Through semistructured interviews with animated 
VR story creators, the thesis identifies ten common stages in the VR creation 
process and highlights nine unique obstacles, thereby informing research 
directions for developing computational approaches tailored to VR. The second 
part explores whether it is feasible to leverage desktop data and knowledge to 
support VR content creation. By focusing on sketches and their time-varying 
stroke gestures, the study demonstrates how commonalities and differences 
between desktop and VR environments can be used to generate VR data from 
desktop data. Finally, the third part further investigates how to adapt in the 
context of color design tasks. The research designs algorithms for color 
extraction, recommendation, and personalization, and develops corresponding 
desktop and VR user interfaces. The findings offer strategies to utilize 
commonalities and differences for adapting desktop computational approaches to 
VR.

The contributions of this thesis include empirical insights into VR content 
creation processes and challenges, the development of algorithms that lower the 
barriers for both desktop and VR content creation, and the design of 
interactive tools tailored to the characteristics of desktop and VR 
environments. By bridging desktop and VR, these efforts can help enrich the VR 
ecosystem with diverse and compelling content, ultimately accelerating the 
adoption of VR technology.


Date:                   Monday, 19 August 2024

Time:                   2:00pm - 4:00pm

Venue:                  Room 3494
                        Lifts 25/26

Chairman:               Prof. Hongbo FU (EMIA)

Committee Members:      Prof. Huamin QU (Supervisor)
                        Prof. Pedro SANDER
                        Dr. Dan XU
                        Dr. Wenhan LUO (EMIA)
                        Prof. Kwan-Liu MA (UC Davis)