A Study on Deep-Learning based In-place Locomotion Technique in Virtual Reality using Multimodal Data Pipeline
- 주제(키워드) Virtual Reality , User experience , Deep learning , Locomotion
- 주제(DDC) 006.31
- 발행기관 아주대학교
- 지도교수 유종빈
- 발행년도 2022
- 학위수여년월 2022. 8
- 학위명 석사
- 학과 및 전공 일반대학원 인공지능학과
- 실제URI http://www.dcollection.net/handler/ajou/000000032188
- 본문언어 영어
- 저작권 아주대학교 논문은 저작권에 의해 보호받습니다.
초록/요약
Movement is one of the key elements in virtual reality (VR) and significantly influences user experience. In particular, walking-in place is a method of supporting movement in a limited space, and many studies are being conducted on its effective support. However, most studies have focused on forward movement despite many situations in which backward movement is needed. In this paper, we present the development of a prediction model for forward/backward movement while considering a user’s orientation and the verification of the model’s effectiveness. We built a deep learning-based model by collecting sensor data through virtual data pipeline which contains a user’s head, waist, and feet. The study was conducted through two technical elements: a data pipeline for collecting signals that could represent a user and a prediction model for a user movement. We developed three realistic VR scenarios that involve backward movement, set three conditions (controller-based, treadmill-based, and model-based) for movement, and evaluated user experience in each condition through a study of 36 participants. As a result, the model-based condition showed the highest sensory sensitivity, effectiveness, and satisfaction and similar cognitive burden compared with the other two conditions. The results of our study demonstrated that movement support through modeling is possible, suggesting its potential for use in many VR applications.
more목차
Chapter 1. Introduction 1
Chapter 2. Related Works 4
Chapter 2.1. Supporting immersive experience in VR 4
Chapter 2.2. Body movement data used in VR 4
Chapter 2.3. Locomotion research in VR 5
Chapter 2.4. Analysis and modeling of multimodal sensor data in VR 7
Chapter 2.5. Learning and predicting movement in VR 8
Chapter 3. Study Procedure 10
Chapter 4. STUDY #1: Backward Movement Detection 11
Chapter 4.1. Study Procedure 11
Chapter 4.2. Data pipeline development 12
Chapter 4.3. Model feature engineering 14
Chapter 4.4. Model development 15
Chapter 4.5. Results 16
Chapter 5. STUDY #2: Effectiveness of Backward Movement Detection Model 18
Chapter 5.1. VR scenarios 18
Chapter 5.2. Methods 20
Chapter 5.2.1. Independent variables 20
Chapter 5.2.2. Dependent variables 21
Chapter 5.3. Study procedure 22
Chapter 5.4. Results 24
Chapter 5.4.1. System log analysis 24
Chapter 5.4.2. User experience analysis 25
Chapter 5.4.3. Interview analysis 28
Chapter 6. DISCUSSION 30
Chapter 6.1. Movement data management 30
Chapter 6.2. Reflection on model development and application 31
Chapter 6.3. Limitations and future work 32
Chapter 7. CONCLUSION 34
Bibliography 35
