Ulrich Engelke

Psychophysiology-Based QoE Assessment: A Survey

We present a survey of psychophysiology-based assessment for quality of experience (QoE) in advanced multimedia technologies. We provide a classification of methods relevant to QoE and describe related psychological processes, experimental design considerations, and signal analysis techniques. We summarize multimodal techniques and discuss several important aspects of psychophysiology-based QoE assessment, including the synergies with psychophysical assessment and the need for standardized experimental design. This survey is not considered to be exhaustive but serves as a guideline for those interested to further explore this emerging field of research.

Varying Spread Fuzzy Regression for Affective Quality Estimation

Design of preferred products requires affective quality information which relates to human emotional satisfaction. However, it is expensive and time consuming to conduct a full survey to investigate affective qualities regarding all objective features of a product. Therefore, developing a prediction model is essential in order to understand affective qualities on a product. This paper proposes a novel fuzzy regression method in order to predict affective quality and estimate fuzziness in human assessment, when objective features are given. The proposed fuzzy regression also improves on traditional fuzzy regression that simulate only a single characteristic with the resulting limitation that the amount of fuzziness is linear correlated with the independent and dependent variables. The proposed method uses a varying spread to simulate nonlinear and nonsymmetrical fuzziness caused by affective quality assessment. The effectiveness of the proposed method is evaluated by two very different case studies, affective design of an electric iron and image quality assessment, which involve different amounts of data, varying fuzziness, and discrete and continuous data. The results obtained by the proposed method are compared with those obtained by the state of art and the recently developed fuzzy regression methods. The results show that the proposed method can generate better prediction models in terms of three fuzzy criteria, which address both predictions of magnitudes and fuzziness.

Perceived interest and overt visual attention in natural images

Region of interest (ROI) based image and video processing has attracted increased research efforts in recent years. The concept of perceptual ROI, however, is not always clearly defined leading to different interpretations between researchers related to bottom-up saliency (signal driven visual attention), top-down attention (subconscious, driven by higher cognitive factors, e.g. interest) or perceived interest. While all of these concepts are likely meaningful in the context of perceptual ROI based image and video processing, it is worth understanding how they are linked altogether. In this paper, the relationship between perceived interest and overt visual attention (which can cover both bottom-up and top-down attention) is studied. Towards this goal, a dedicated ROI selection experiment was performed and is analysed in detail, revealing deep insight into perceived interest in natural images. The outcomes are compared to an eye gaze tracking experiment representing overt visual attention in natural images. It is shown that there is indeed a strong relationship between perceived interest and overt visual attention for a wide range of natural scenes. We show that the relationship has a strong dependence on image content and presentation time during the eye gaze tracking experiment. Furthermore, eye gaze tracking data is revealed to have a high predictive value of primary ROI, particularly in case of the latter dominating over the remainder of the image. Both, the ROI and the eye gaze tracking databases are made publicly available to the research community. HighlightsStudy on the relationship between overt visual attention and perceived interest.Eye gaze tracking and ROI selection experiments as solid ground truths.Effects of image content and image presentation time are investigated in depth.Quantitative analysis shows strong relationship in images with dominant ROI.The data of both experiments is made publicly available to the research community.

MelissAR: Towards Augmented Visual Analytics of Honey Bee Behaviour

We present the design and current prototype implementation of MelissAR, an augmented reality system for visual analytics of honey bee behaviour in the field. The system is intended to support bee keepers and other relevant users to monitor honey bee populations and to make effective decisions based on their status. The implementation of MelissAR is based on informed design choices with regard to usability in the field, effective communication of relevant information, and robustness to varying outdoor conditions.

Review on using physiology in quality of experience.

In the area of Quality of Experience (QoE), one challenge is to design test methodologies in order to evaluate the perceived quality of multimedia content delivered through technical systems. Traditionally, this evaluation is done using subjective opinion tests. However, sometimes it is difficult for observers to communicate the experienced quality through the given scale. Furthermore, those tests do not give insights into how the user is reacting on an internal physiological level. To overcome these issues, one approach is to use physiological measures, in order to derive a direct non-verbal response of the recipient. In this paper, we review studies that have been performed in the domain of QoE using physiological measures and we look into current activities in standardization bodies. We present challenges this research faces. and give an overview on what researchers should be aware of when they want to start working in this research area.

Visual and Manual Control for Human-Robot Teleoperation

In an effort to assess simple control modalities for a remote robotic system, this study explores and tests the suitability of four interfaces for teleoperation in human-robot interaction. For a pick-and-place task, users were asked to select targets and locations using eye tracking (activated by either a mouse click or dwell time), a touchscreen, or a standard computer mouse. Contrary to their expectations, the authors found that eye-tracking-based interaction, especially when paired with manual-click selection, was generally slower and was perceived as more difficult than the mouse and touchscreen interfaces. Conversely, as predicted, they found evidence that eye tracking with dwell selection was less prone to interference caused by a secondary manual task.

A Visual Analytics Framework to Study Honey Bee Behaviour

We present an end-to-end visual analytics framework that aims to facilitate prediction and decision making about honey bee health based on micro sensing data. The framework is particularly tailored to cope with heterogeneous data from micro sensors and environmental sensors that are deployed to collect information about bees and their environment. The framework design allows for a wide range of end users, including scientists, bee keepers, and decision makers, to effectively explore the bee data through interactive visual interfaces. User centred design is deployed throughout the development to meet the various requirements of the users. A large scale study is being planned to evaluate and further refine the framework based on user experiences.

Design of Environmental Sensor Networks Using Evolutionary Algorithms

An evolutionary algorithm (EA)-assisted spatial sampling methodology is proposed to assist decision makers in sensor network (SN) deployments. We incorporated an interpolation technique with leave-one-out cross-validation (LOOCV) to assess the representativeness of a particular SN design. For the validation of our method, we utilized Tasmanias South Esk Hydrological Model developed by the Commonwealth Scientific and Industrial Research Organisation, which includes a range of environmental variables describing the landscape. We demonstrated that our proposed methodology is capable of assisting in the initial design of SN deployment. Ordinary Kriging is shown to be the best suited spatial interpolation algorithm for the EAs LOOCV under the current empirical study.

Collaborative Framework Design for Immersive Analytics

Recent trends in computing environments indicate that the future infrastructure for visual analytics will be distributed and collaborative. Collaborative frameworks create value for scientists, analysts, industrial partners, domain experts, and other end-users to meet, communicate, interact with others, and coordinate their activities in a globally shared network. This paper focuses on collaborative framework design for immersive analytics facilitating the integration of multimodal immersive interfaces. The framework design takes into account visualisation and interaction techniques for multiple users and especially decision support tools for scientific visual analytics experts. An overview of several important aspects of collaborative platforms for immersive analytics is presented and different modules of our proposed platform (including data management, analytics, visualisation, querying, and user interface design) will be detailed to highlight their importance in a full visual analytics pipeline.

Fuzzy regression for perceptual image quality assessment

Subjective image quality assessment (IQA) is fundamentally important in various image processing applications such as image/video compression and image reconstruction, since it directly indicates the actual human perception of an image. However, fuzziness due to human judgment is neglected in current methodologies for predicting subjective IQA, where the fuzziness indicates assessment uncertainty. In this article, we propose a fuzzy regression method that accounts for fuzziness introduced through human judgment and the limitations of widely-used psychometric quality scales. We demonstrate how fuzzy regression models provide fuzziness information regarding subjective IQA. We benchmark the fuzzy regression method against the commonly used explicit modeling method for subjective IQA namely statistical regression by considering three real situations involving subjective image quality experiments where: (a) the number of participants is insufficient; (b) an insufficient amount of data is used for modelling; and (c) variant fuzziness is caused by human judgment. Results indicate that fuzzy regression models achieve more effective data fitting and better generalization capability when predicting subjective IQA under different types and levels of image distortion.