Omnidirectional videos (ODVs, or 360o videos) have been studied extensively in recent years. There is a vast number of algorithms and devices to capture, construct, and display omnidirectional video content, and large enterprises including Vimeo and YouTube offer their own platforms for viewing these videos. Omnidirectional videos have been used, for instance, in remote operations and telepresence applications (see, for example, publications VI and VII; Onoe et al. 1998; Boult, 1998). They have also been employed to supply immersive experiences to users in cultural context, say, in museums (Kwiatek and Woolner, 2010) and theaters (Decock et al., 2014). Other domains where ODVs have been utilized include education, for example, in teaching secondary languages (Publication VII) and sign language (Järvinen and Ekola, 2014).
One interesting field of research in which ODVs have been used recently is health care. VEs have been useful tools for studying and treating patients, for which the term virtual reality exposure theory, or VRET (Riva, Botella, Légeron and Optale, 2004) has been adopted. There is some evidence that people with higher levels of anxiety report higher levels in presence (Alsina-Jurnet, Gutiérrez-Maldonado, & Rangel-Gómez, 2011). This also makes them experience greater anxiety when they are exposed to phobia- or fear-inducing stimuli within the VE. These findings support the notion that VEs, including those that use ODVs, are useful tools for studying and treating phobias and fears. Fassbender and Heiden (2014) implemented an application, Atmosphaeres, for stress and pain management, and Rizzo et al. theorized about practical uses for ODVs in therapy.
One aspect of applications that utilize ODVs is their interactivity. In addition to viewing their video content, one can also add other interactive elements to them. For example, the user can navigate video content or gather contextual information about the environment presented in the content. For these applications, we invented the term interactive ODVs, or
……………
55
iODVs for short (Publication VI). Guidelines for designing these applications have been suggested by Saarinen et al. (2017) and Argyriou et al. (2016). Multimodality and interaction have also been studied in the context of ODVs and, for example, gesture-based interaction (Rovelo-Ruiz, 2014) and second screen interfaces (Zoric et al., 2013) have been used for interacting with ODV content. Benko and Wilson (2010) implemented an application that utilizes ODVs, has a multi-user support, and supports mid-air gestures. In publications VI and VII, we presented an application that utilizes position-based interaction with a dwell timer for HMD applications, and for CAVE systems, we developed an interaction method that employs a rotating chair with a built-in rotation sensor.
Collaboration within VEs that utilize ODV content have also been studied in recent years. Singhal and Neustaedter (2017) developed an application, BeWithMe, that allows long-distance couples to collaborate and communicate. In this application, the users can share ODVs about their daily life and experiences. The use of ODVs in a shared guided tour was evaluated by Tang and Fakourfar (2017). Participants in their study had difficulties in building “a shared understanding of what was being looked at and discussed,” which might be due to the low interactivity of the application. Ramalho and Chambel (2013) simulated wind to enhance the subject’s experience with ODVs. This type of multisensory augmentation is another potential research subject related to ODVs.
4.5 SUMMARY
This chapter explained the history and basic concepts of VEs and VR systems. The first virtual reality experiments date back to the 1800s, when Wheatstone experimented with the use of stereoscopic images. This again led to the development of the lenticular stereoscope in 1849 and the View-Master in 1939. These devices applied the same principles that modern HMD devices use. The first VEs connected to a computer were developed during the 1960s. In addition, a short introduction on collaborative VEs and VLEs was presented.
Subsequently, this researcher summarized the background of and work related to wayfinding in VEs. The basic problems were stated concerning the use of pen-and-paper evaluations and self-reported measures for wayfinding studies. In addition, it was suggested that using traditional interaction methods such as a keyboard and mouse may not provide researchers with comprehensive results on wayfinding, as the motor, vestibular, and proprioceptive systems are not utilized like they usually are when traveling in real large-scale environments.
Guidelines were provided for designing prominent landmarks that support wayfinding for VEs for both urban and natural environments, and it was
suggested that VEs using photorealistic environments may supply a more realistic experience for wayfinding than three-dimensional models. Gender differences were also discussed regarding wayfinding in VEs, including the possible reasons for differences in wayfinding performance between females and males. Then immersion and presence were described, alongside concepts related to these phenomena. I will discuss this terminology further in the Discussion chapter of this thesis. Finally, research was mentioned concerning ODVs and iODVs, including potential research topics for wayfinding studies with these content types.
……………
57
5 Introduction to the Publications
The research for this dissertation consists of designing and evaluating a model for highlighting landmarks for pedestrian wayfinding, as well as implementing applications for collaborative wayfinding in VEs and the user studies conducted with them. The research articles presented in this dissertation target the following topics:
This paper presented a model for highlighting landmarks in pedestrian wayfinding applications (Publication I).
This work assessed the model developed in the previous publication (Publication II).
These articles explored the user experience with and immersion in collaborative VEs (publications III, IV, and V).
This work evaluated the sensation of immersion and user experience with iODVs between CAVE systems and HMDs (Publication VI).
Interactive ODVs (Publication VII): this model was used in the design and implementation of contextual information for the applications.
The switch of focus from landmark-based wayfinding to collaborative wayfinding in VEs came naturally because the model for landmark highlighting (Publication I) that was developed required an application for evaluating it. After reading the related work for pedestrian wayfinding, it was evident that there was still a large gap in research in collaborative wayfinding and that the model developed earlier could be utilized in this research. When sufficient realism and fidelity are provided, the transfer knowledge between VEs and real-world situations are comparable (Witmer
et al., 1996). Waller et al. (1998) suggested that, because of the variability of VEs, some training scenarios can be even superior to real-world setups.
The first application that utilized the model was Berlin Kompass (Publication II), which created contextual information that supported the collaborative wayfinding task.
In Publication III, this concept of collaborative wayfinding was utilized in the context of language learning. The results suggested that this concept has pedagogical potential, but also stated its clunky, complex installation might be a hindrance for its actual use in pedagogical settings.
After this, the same application was utilized for evaluating the influence of gender and game experience on user behavior (Publication IV).
Subsequently, the clunky setup of Berlin Kompass was developed further into a web version that employed modern web technologies and allowed users to improve their wayfinding and language skills with just a web browser. This application, CityCompass (Publication V) was much easier to set up and use in various environments.
Publication VI compared the feeling of immersion and user experience between two VEs, HMD and CAVE. This aim of the study was to detect any differences between the two media and give guidance on which platform would be more suitable for future applications.
For Publication VII, a CityCompass VR application utilizing iODVs and HMD was developed. This paper studied variations in the sensation of immersion between genders while interacting in collaborative VEs. The following sections will explore and explain each publication in greater detail, starting with Publication 1.
5.1 MODEL FOR LANDMARK HIGHLIGHTING IN MOBILE WEB SERVICES