Towards an Effec�ve Arousal Detec�on System for Virtual Reality I�igeneia Mavridou Centre of Digital Entertainment Bournemouth University Bournemouth, UK imavridou@bournemouth.ac.uk Ellen Seiss Department of Psychology Bournemouth University Bournemouth, UK eseiss@bournemouth.ac.uk Theodoros Kostoulas Dept. of Computing and Informatics Bournemouth University Bournemouth, UK tkostoulas@bournemouth.ac.uk Charles Nduka Emteq Ltd. Science Park Square, Brighton, UK charles@emteq.net Emili Balaguer-Ballester Dept. of Computing and Informatics Bournemouth University Bournemouth, UK eb-ballester@bournemouth.ac.uk ABSTRACT Immersive technologies offer the potential to drive engage- ment and create exciting experiences. A better understanding of the emotional state of the user within immersive experi- ences can assist in healthcare interventions and the evaluation of entertainment technologies. This work describes a feasibil- ity study to explore the effect of affective video content on heart-rate recordings for Virtual Reality applications. A low- cost re�lected-mode photoplethysmographic sensor and an electrocardiographic chest-belt sensor were attached on a novel non-invasive wearable interface specially designed for this study. 11 participants responses were analysed, and heart-rate metrics were used for arousal classi�ication. The reported results demonstrate that the fusion of physiological signals yields to signi�icant performance improvement; and hence the feasibility of our new approach. CCS CONCEPTS •Human-centered computing→ Interaction paradigms: Virtual reality; •Information systems→ Sentiment analysis; • Human-centered computing→ Interactive systems and tools KEYWORDS Virtual Reality; Arousal, Classification; PPG; ECG; C-SVM; ACM Reference format: Ifigeneia Mavridou, Ellen Seiss, Theodoros Kostoulas, Charles Nduka, Emili Balaguer-Ballester. 2018. Towards an Effective Arousal Detec- tion System for Virtual Reality. In Proc. of ACM Human-Habitat for Health (H3'18). ACM, Boulder, CO, USA, October 2018, 6 pages. DOI: 10.1145/3279963.3279969 Permission to make digital or hard copies of part or all of this work for personal or classroom use is granted without fee provided that copies are not made or distributed for profit or commercial advantage and that copies bear this notice and the full citation on the first page. Copyrights for components of this work owned by others than ACM must be honored. Abstracting with credit is permit- ted. To copy otherwise, or republish, to post on servers or to redistribute to lists, requires prior specific permission and/or a fee. Request permissions from Permissions@acm.org. Human-Habitat for Health (H3'18), October 16, 2018, Boulder, CO, USA © 2018 Association for Computing Machinery. ACM ISBN 978-1-4503-6075-3/18/10…$15.00 https://doi.org/10.1145/3279963.3279969 1 INTRODUCTION The increasingly evolving Virtual Reality (VR) technologies permit the adaptation of experimental protocols for their use with VR. Crucially, experiment design utilising VR can offer controlled laboratory conditions while granting a wealth of content resources and ecological validity [1]. User input and interface sensory modalities are currently integrated with VR, as they monitor the user’s actions. These systems use various haptic and wearable user-interfaces to track head and body movements, eye gaze and speech patterns [2]. Such metrics can describe useful information related to the user’s behav- iour, preferences and actions within VR. As such, they can im- prove automatic emotion recognition, which is important to enhance VR user interactions. Previous research on affective computing offers a wealth of emotion detection solutions ranging from physiological and speech signals, to monitoring facial expressions, and movement analysis [3]. Understanding the user's emotions and behaviour within VR experiences could not only assist experience-designers to evaluate their content [4, 5] but also in healthcare interventions such as VR exposure therapy [6]. There are two basic challenges for emotion recognition in VR. Firstly, the Head Mounted Displays (HMDs) commonly used during VR experiences cover a significant part of the face which renders the detection of facial expressions difficult. Secondly, commercial immersive experiences require often intense head and limb movements, which could result in noise artefacts on potential wearable sensors. To overcome the first challenge, our team developed a novel prototype for facial ex- pression recognition, Faceteq™ [7] with surface physiological sensors. This interface can be incorporated on a commercial HMD, acting as non-invasive, soft medium between the user’s skin and the HMD. In this work, we propose a system for the detection of high and low arousal in VR settings via capturing multimodal heart-rate responses (from low cost, custom-made photople- thysmographic (PPG) and electrocardiographic (ECG) sen- sors) and continuous self-ratings of HMD users. brought to you by CORE View metadata, citation and similar papers at core.ac.uk provided by Bournemouth University Research Online