STUDY ON DEEP LEARNING MODELS FOR THE CLASSIFICATION OF VR SICKNESS LEVELS
Virtual Reality (VR) sickness is often accompanied by symptoms such as nausea and dizziness, and a prominent theory explaining this phenomenon is the sensory conflict theory. Recently, studies have used Deep Learning to classify VR sickness levels; however, there is a paucity of research on Deep Le...
Saved in:
| Main Authors: | , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
Polish Association for Knowledge Promotion
2024-12-01
|
| Series: | Applied Computer Science |
| Subjects: | |
| Online Access: | https://ph.pollub.pl/index.php/acs/article/view/6474 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1841552801994375168 |
|---|---|
| author | Haechan NA Yoon Sang KIM |
| author_facet | Haechan NA Yoon Sang KIM |
| author_sort | Haechan NA |
| collection | DOAJ |
| description |
Virtual Reality (VR) sickness is often accompanied by symptoms such as nausea and dizziness, and a prominent theory explaining this phenomenon is the sensory conflict theory. Recently, studies have used Deep Learning to classify VR sickness levels; however, there is a paucity of research on Deep Learning models that utilize both visual information and motion data based on sensory conflict theory. In this paper, the authors propose a parallel merging of a Deep Learning model (4bay) to classify the level of VR sickness by utilizing the user's motion data (HMD, controller data) and visual data (rendered image, depth image) based on sensory conflict theory. The proposed model consists of a visual processing module, a motion processing module, and an FC-based VR sickness level classification module. The performance of the proposed model was compared with that of the developed models at the time of design. As a result of the comparison, it was confirmed that the proposed model performed better than the single model and the merged (2bay) model in classifying the user's VR sickness level.
|
| format | Article |
| id | doaj-art-d8df541b5c4947c5a09958006fb93070 |
| institution | Kabale University |
| issn | 2353-6977 |
| language | English |
| publishDate | 2024-12-01 |
| publisher | Polish Association for Knowledge Promotion |
| record_format | Article |
| series | Applied Computer Science |
| spelling | doaj-art-d8df541b5c4947c5a09958006fb930702025-01-09T12:44:47ZengPolish Association for Knowledge PromotionApplied Computer Science2353-69772024-12-0120410.35784/acs-2024-37STUDY ON DEEP LEARNING MODELS FOR THE CLASSIFICATION OF VR SICKNESS LEVELSHaechan NA0Yoon Sang KIM1https://orcid.org/0000-0002-0416-7938Korea University of Technology and EducationKorea University of Technology and Education, Institute for Bioengineering Application Technology, Department of Computer Science and Engineering, BioComputing Lab Virtual Reality (VR) sickness is often accompanied by symptoms such as nausea and dizziness, and a prominent theory explaining this phenomenon is the sensory conflict theory. Recently, studies have used Deep Learning to classify VR sickness levels; however, there is a paucity of research on Deep Learning models that utilize both visual information and motion data based on sensory conflict theory. In this paper, the authors propose a parallel merging of a Deep Learning model (4bay) to classify the level of VR sickness by utilizing the user's motion data (HMD, controller data) and visual data (rendered image, depth image) based on sensory conflict theory. The proposed model consists of a visual processing module, a motion processing module, and an FC-based VR sickness level classification module. The performance of the proposed model was compared with that of the developed models at the time of design. As a result of the comparison, it was confirmed that the proposed model performed better than the single model and the merged (2bay) model in classifying the user's VR sickness level. https://ph.pollub.pl/index.php/acs/article/view/6474VR sicknessCyber sicknessDeep LearningLSTMResNetSensory Conflict Theory |
| spellingShingle | Haechan NA Yoon Sang KIM STUDY ON DEEP LEARNING MODELS FOR THE CLASSIFICATION OF VR SICKNESS LEVELS Applied Computer Science VR sickness Cyber sickness Deep Learning LSTM ResNet Sensory Conflict Theory |
| title | STUDY ON DEEP LEARNING MODELS FOR THE CLASSIFICATION OF VR SICKNESS LEVELS |
| title_full | STUDY ON DEEP LEARNING MODELS FOR THE CLASSIFICATION OF VR SICKNESS LEVELS |
| title_fullStr | STUDY ON DEEP LEARNING MODELS FOR THE CLASSIFICATION OF VR SICKNESS LEVELS |
| title_full_unstemmed | STUDY ON DEEP LEARNING MODELS FOR THE CLASSIFICATION OF VR SICKNESS LEVELS |
| title_short | STUDY ON DEEP LEARNING MODELS FOR THE CLASSIFICATION OF VR SICKNESS LEVELS |
| title_sort | study on deep learning models for the classification of vr sickness levels |
| topic | VR sickness Cyber sickness Deep Learning LSTM ResNet Sensory Conflict Theory |
| url | https://ph.pollub.pl/index.php/acs/article/view/6474 |
| work_keys_str_mv | AT haechanna studyondeeplearningmodelsfortheclassificationofvrsicknesslevels AT yoonsangkim studyondeeplearningmodelsfortheclassificationofvrsicknesslevels |