Publikace

A navigation system for visually impaired users can be much more efficient if it is based on collaboration among visually impaired persons and on utilising distributed knowledge about the environment in which the navigation task takes place. To design a new system of this kind, it is necessary to make a study of communication among visually impaired users while navigating in a given environment and on their regularly walked routes. A qualitative study was conducted to gain insight into the issue of communication among visually impaired persons while they are navigating in an unknown environment, and our hypotheses were validated by a quantitative study with a sample of 54 visually impaired respondents. A qualitative study was conducted with 20 visually impaired participants aimed at investigating regularly walked routes used by visually impaired persons. The results show that most visually impaired users already collaborate on navigation, and consider an environment description from other visually impaired persons to be adequate for safe and efficient navigation. It seems that the proposed collaborative navigation system is based on the natural behaviour of visually impaired persons. In addition, it has been shown that a network of regularly walked routes can significantly expand the urban area in which visually impaired persons are able to navigate safely and efficiently.

User interfaces (UI) of software applications play a crucial part in communication with users. Attractive UIs often lead to market success, and thus there is a significant incentive to provide users with malleable UIs that can adapt as much as possible to their needs. However, such UIs require significant development and maintenance efforts. In this paper, we describe a context model based on ability-based design that is well suited to the purposes of automated UI generation. We then introduce a platform that delivers adaptive UIs across various platforms. We use runtime combinatoric optimisation to support usability and to generate context-sensitive UIs. Since the development and maintenance of such UIs can be complex, our platform integrates a module for code-inspection for data-oriented applications to reduce these efforts. It also utilises a visual editor to simplify manual UI design.

User behavior is significantly influenced by the surrounding environment. Especially complex and dynamically changing environments (like mobile environment) are represented by a wide variety of extraneous variables, which influence the user behavior in an unpredictable and mostly uncontrolled way. For researchers, it is challenging to measure and analyze the user behavior in such environments. We introduce a complex tool-the IVE tool-which provides a unique way of context visualization and synchronization of measured data of various kinds. Thanks to this tool it is possible to efficiently evaluate data acquired during complex usability tests in a mobile environment. The functionality of this tool is demonstrated on the use case "Navigation of visually impaired users in the building with support of a navigation system called NaviTerier." During the experiment, we focused on collection and analysis of data that may show user stress and which may influence his/her ability to n

Turn-taking and turn-yielding phenomena in dialogs receive increasing attention nowadays. A growing number of spoken dialog systems inspire application designers to humanize people’s interaction experience with computers. The knowledge of psychology in discourse structure could be helpful in this effort. In this paper the authors explore effectiveness of selected visual and vocal turn-yielding cues in dialog systems using synthesized speech and an avatar. The aim of this work is to detect the role of visual and vocal cues on dialog turn-change judgment using a conversational agent. The authors compare and study the cues in two experiments. Findings of those experiments suggest that the selected visual turn-yielding cues are more effective than the vocal cues in increasing correct judgment of dialog turn-change. Vocal cues in the experiment show quite poor results and the conclusion discusses possible explanations of that.

We propose a novel method for computing labeling of 3D illustrations in real-time. We solve a multiple criteria optimization problem in which we consider the desired layout already in the stage of searching for salient points of the labeled areas. In the solution we employ fuzzy logic combined with greedy optimization. The method runs on the GPU and achieves interactive rates on medium sized models. The results indicate that the method compares favorably to the state-of-the-art interactive labeling techniques.

We present an external labeling laid over small and/or overlapping 2D objects as an efficient representation for their selection. The approximation of objects with points allows us to transform the labeling problem to graph layout problem, which we solve by means of force-based algorithm. The input parameters allow us to influence the resulting layout of label boxes (e.g. to adapt their distance for imprecise input devices). In a study with 15 participants two implementations of our algorithm were compared against labeling method, where all label boxes share the same offset from corresponding objects. The results of the study show that implementation using a special functionality (temporary freezing of the label box position recalculation) was 14% faster with a comparable accuracy. The subjective evaluation revealed that the implementation with temporary freezing is perceived as most comfortable, fastest and most accurate. The implementation without temporary freezing showed much higher error rate and cannot be recommended.