Matevž Pogačnik

An Analysis of the Precision and Reliability of the Leap Motion Sensor and Its Suitability for Static and Dynamic Tracking

We present the results of an evaluation of the performance of the Leap Motion Controller with the aid of a professional, high-precision, fast motion tracking system. A set of static and dynamic measurements was performed with different numbers of tracking objects and configurations. For the static measurements, a plastic arm model simulating a human arm was used. A set of 37 reference locations was selected to cover the controllers sensory space. For the dynamic measurements, a special V-shaped tool, consisting of two tracking objects maintaining a constant distance between them, was created to simulate two human fingers. In the static scenario, the standard deviation was less than 0.5 mm. The linear correlation revealed a significant increase in the standard deviation when moving away from the controller. The results of the dynamic scenario revealed the inconsistent performance of the controller, with a significant drop in accuracy for samples taken more than 250 mm above the controllers surface. The Leap Motion Controller undoubtedly represents a revolutionary input device for gesture-based human-computer interaction; however, due to its rather limited sensory space and inconsistent sampling frequency, in its current configuration it cannot currently be used as a professional tracking system.

Intuitive gesture based user identification system

We present an intuitive and very easy to use implicit gesture based identification system that is especially suited for security-wise non-critical applications, such as the user login in the multimedia services. The term “implicit gesture” in our work refers to a natural physical hand manipulation performed by the user who takes hold of the control device and simply picks it up from its neutral motionless position - a “pick-up” gesture. For reference with other related systems, explicit and well defined personal name identification gestures were used as well. User evaluation study results show that 100% recognition accuracy was achieved when using explicit identification signature gestures and over 91% recognition accuracy was achieved when using the implicit “pick-up” gesture. Performance of the proposed system is comparable to results of other respectable related works when using explicit identification gestures, while also showing that implicit gesture based user identification is possible and viable.

A Presence-Based Context-Aware Chronic Stress Recognition System

Stressors encountered in daily life may play an important role in personal well-being. Chronic stress can have a serious long-term impact on our physical as well as our psychological health, due to ongoing increased levels of the chemicals released in the ‘fight or flight’ response. The currently available stress assessment methods are usually not suitable for daily chronic stress measurement. The paper presents a context-aware chronic stress recognition system that addresses this problem. The proposed system obtains contextual data from various mobile sensors and other external sources in order to calculate the impact of ongoing stress. By identifying and visualizing ongoing stress situations of an individual user, he/she is able to modify his/her behavior in order to successfully avoid them. Clinical evaluation of the proposed methodology has been made in parallel by using electrodermal activity sensor. To the best of our knowledge, the system presented herein is the first one that enables recognition of chronic stress situations on the basis of user context.

MedReminder - an interactive multimedia medical application for the IPTV environment

We present the MedReminder, an interactive multimedia medical application, integrated into an existing IPTV service framework. The MedReminder application is implemented within XBMC multimedia platform and is integrated alongside interactive TV live and on demand multimedia services, providing easy to use and intuitive user experience. It is a client-server based application connecting different target user roles: doctors as medicine prescribers, patients as medicine consumers and patient caregivers. It allows for a centralized collection of medicine related reminder information. Consequently, the system provides user notifications on TV screen and through other communication channels, such as SMS messages, e-mail and telephone calls, which are used in case the consumption of prescribed medicine has not been confirmed. Several use case scenarios are discussed including the possibility of dynamic medicine prescriptions, based on doctors remote readings of patients medical parameters (e.g. blood pressure, body weight, temperature etc.), using the integrated personal health system.

An automatic Wi-Fi-based approach for extraction of user places and their context

With the analysis of various sensor data from the mobile devices, it is possible to extract user situations, so-called user context. This is needed for the development of modern, user-friendly services. Therefore, we developed a simple, nonintrusive, and automatic method based on the Wi-Fi fingerprints and GPS. The method finds user stay points, aggregates them into meaningful stay regions, and assigns them four general user contexts: home, work, transit, and free time. We evaluated its performance on the real traces of six different users who annotated their contexts over eight days. The method determined the stay mode of the users with accuracy, precision, and recall of above 96%. In combination with the novel approach for aggregation, all regions relevant to the users were determined. Among the tested aggregation schemes, the fingerprint similarity approach worked the best. The context of the determined stay regions was on average accurately inferred in 98% of the time. For the contexts home, work, and free time, the precision and recall exceeded 86%. The results indicate that the method is robust and can be deployed in various fields where context awareness is desired.

Capturing the mood: Evaluation of the moodstripe and moodgraph interfaces

This study presents an evaluation of two interfaces for gathering user feedback in online surveys. We evaluated the intuitiveness, usability and time complexity of the proposed interfaces in comparison to the more standard approaches. Over 900 users first participated in an online survey exploring the influence of mood on their emotional responses to music and colors. We included several new interfaces in this survey, so after it was completed, users were asked to complete a second survey where they evaluated various aspects of the interfaces. Our analysis shows reduced time complexity and increased intuitiveness of the new interfaces, compared to standard approaches, resulting in lower mental difficulty and frustration of participants.

Users’ viewpoint of usability and user experience testing procedure - gaining methodological insights in a case of an interactive HbbTV application

We present a novel meta-methodological approach for the user experience and usability methodology procedure evaluation, shown in an example of the user experience and usability study of an interactive HbbTV application. The idea behind this research is not only to evaluate and improve the TV-WEB service but also to gain insights how the participants perceived the whole Ux evaluation procedure itself. The research questions focused mainly on the time complexity (temporal demand) and the frustration level of the TV-WEB evaluation procedure. Additionally the appropriateness of the selected content used, interface and interaction design, and the service impressions and satisfaction/payment related questions was sought. A special questionnaire based on the NASA TLX standard test is presented. The concept has been successfully implemented in several live field trials in three countries (BiH, Serbia and Montenegro). In the user experience and usability study of the service itself more than 150 participants were involved, of those 35 took part in the meta-methodology study. The feedback was quantitatively evaluated on a 7-point Likert scale, with “1” indicating the best positive feedback, “4” neutral/undecided feedback and “7” the most negative feedback. The quantitative average summary results obtained in the three evaluation studies were 1.30 for both the BiH and Serbia test study cases and the score of 1.22 for the test study case in Montenegro. These results show that a great majority of the participants found the whole evaluation procedure time-wise and frustration-wise undemanding, with appropriate content, presentation style and overall attitude towards them. By using this approach it was possible to improve the user experience and usability methodology used, producing more reliable results and providing better user experience in the final version of the product as well as providing a pleasant experience during the testing of the product.

CONTENT AND PRESENTATION ADAPTATION IN HYPERMEDIA SYSTEMS

The present chapter is dedicated to personalised real-time multimedia communication services. The challenge of modern communication services is a combination of personalised information gathering and customised data retrieval technology. A concept of personalised communication service is presented, followed by an overview of information search, retrieval and adaptive presentation technologies. Both generalised strategies and supporting technologies are presented throughout the chapter.

Smart Driving: Influence of Context and Behavioral Data on Driving Style

In this article, we present an approach to determine stress level in a non-invasive way using a smartphone as the only and sufficient source of data. We also present the idea of how to partly transfer such approach to the determination of the driving style, as aggressive driving is one of the causes of car accidents. For determination of the driving style a variety of methods are used including the preparation movements before maneuvers, identification of steering wheel angle, accelerator and brake pedal pressures, glance locations, facial expressions, speed, medical examinations before driving as well as filling out of the questionnaires after the journey. In our paper we present a methodology for estimation of potentially unsafe driving (in the meaning of more intensive acceleration and braking compared to average driving) and discuss how to estimate such unsafe driving before it actually takes place. We present sensors and data which can be used for these purposes. Such data include heart rate variability from chest belt sensor, behavioral and contextual data from smartphone, STAI short questionnaire to assess personal anxiety and anxiety as a state at certain moment, and initial interaction with car during opening and closing of the car doors. To determine intensive acceleration and braking we analyzed GPS data like speed, acceleration and also data from accelerometer inside the car to avoid interference in GPS-signal. Actually, our long term goals are to provide feedback about potentially unsafe driving in advance and thus strengthening driver’s attention on the driving process before the start.

Influence of video content type on users’ virtual reality sickness perception and physiological response

Abstract Virtual Reality (VR) sickness (Cybersickness) is an affliction and a challenge, common to users of virtual environments. We therefore asked ourselves this research question: “Can video content type influence users’ VR sickness and physiological response?” We conducted a study with 26 participants, who watched two omnidirectional videos of different content types (neutral and action content) on five distinct video conditions: 2D TV screen, three generations of Oculus Rift VR HMDs (DK1, DK2 and CV1) and on the mobile Samsung GearVR HMD. The Simulator Sickness Questionnaire and the Subjective Units of Distress Scale in combination with the measurement of the physiological parameters (electrodermal activity and skin temperature, respiratory frequency and heart rate) were used to assess the VR sickness effects. The results show that video content type as well as users’ background preferences (preference to adrenaline sports) affected the users’ VR sickness perception. Considering various video conditions, significantly less VR sickness effects were reported with the TV condition than with any VR devices. The results of the subjective questionnaires were correlated with the objective physiological measurements, whereby skin conductance strongly correlated with the VR sickness effects. The effects were also more pronounced in cases of action video content type. Furthermore, we show there is a strong correlation when assessing the VR sickness effects using subjective questionnaire-based methods (the Simulator Sickness Questionnaire and the Subjective Units of Distress Scale) of various complexity, indicating the simple methods (only one question), can effectively be used as well.