14.10.2014 |
Institute meeting |
|
|
21.10.2014 |
Opening up Geoinformatics: Tales from my Sabbatical |
Christian Kray |
Institut für Geoinformatik, Universität Münster |
|
In this brief talk, I will report on research and other activities that I carried out during my (all too short) sabbatical last semester. A key theme that emerged during that time is 'openness'. It pertains to several areas including how people interact with geoinformation, how they can be empowered by it and fundamentally, how Geoinformatics itself can become more open. In this talk, I will present some ideas and first steps towards realising openness, and I will also discuss some open issues and interesting directions for future research. |
Freitag 31.10.2014 |
Per Anhalter durch die Geoinformatik (Antrittsvorlesung) |
Albert Remke |
Institut für Geoinformatik, Universität Münster
52°North – Initiative for Geospatial Open Source Software GmbH
Esri Deutschland Group GmbH |
|
Douglas Adams – ein Vertreter der Monty Python Generation – hat 1978 in seinem humorvollen Science Fiction Roman „Per Anhalter durch die Galaxis“ überraschende Perspektiven auf die Menschheit und die Informationstechnologie entwickelt. Perspektivwechsel bringen neue An – und Einsichten und sind als solche ein wichtiges methodisches Element von Forschung und Innovation. Der Vortrag greift einige Aspekte aus dem Roman auf und setzt sie in Beziehung zu Themen der Geoinformatik. |
04.11.2014 |
Fields as a Generic Data Type for Big Spatial Data |
Gilberto Camara |
Institut für Geoinformatik, Universität Münster |
|
One of the biggest changes in Geoinformatics in recent years arises from technologies that produce lots of data. Earth observation and navigation satellites, mobile devices, social networks, and smart sensors create large data sets with space and time references. To deal with these large data sets, we need appropriate conceptual models, that lead to sound implementations. In this work, we introduce the Fields data type as a possible way to represent large data sets.
Most big spatial data sets provide information about properties of reality in continuous way, which leads to their representation as fields. We propose a generic data type for fields that can represent different types of spatiotemporal data, such as trajectories, time series, remote sensing and, climate data. We show how to represent existing algebras for spatial data with the Fields data type. We also argue that array databases are the best support for processing big spatial data and show how to use the Fields data type with array databases.
|
11.11.2014 |
Understanding the semantics of (visual) data: A perspective of pattern recognition and computer vision |
Xiaoyi Jiang |
Universität Münster |
|
Pattern recognition (nearly synonymous with machine learning) is the science of supervised classification of patterns based on learned models and unsupervised recognition of regularities in data. Computer vision is focused on acquiring, processing, analyzing, and understanding images and videos. Both pattern recognition and computer vision provide powerful means for understanding the semantics of (visual) data from a particular perspective, which ultimately enables intelligent decision-making processes. After a brief introduction to these two fields the main focus of this talk is to give an overview of the related research activities in my group. |
18.11.2014 |
COMPUTATIONAL COGNITIVE SYSTEMS - The Case of Commonsense, Space, and Change |
Mehul Bhatt |
Universität Bremen |
|
A particular emphasis of my research initiatives in recent years has been on the formal and computational characterisation of ‘space’ and ‘visuo-spatial’ problem-solving processes (e.g., abstraction, reasoning, learning) within a range of assistive technologies in domains such as cognitive vision and robotics, architecture and urban design, and geographic information systems. The central underlying theme has been that of reasoning about “space, events, actions, change, and interaction”. The motivating question has been: how can expert analytical and commonsense reasoning capabilities, as exhibited by humans in specific problem contexts, be used as a model and benchmark for the development of collaborative cognitive interaction systems (concerned with people assistance, assurance, and empowerment)?
In this talk, I will formally and computational characterise visuo-spatial “perceptual sense-making processes” that mediate sensing capability / sensory data, and the formation of sensible impressions (e.g., abstractions, analogies, hypotheses and theory formation, beliefs and their revision, argument formation) in expert domain-specific problem solving (e.g., spatial design, geospatial analysis), or even in regular activities of everyday living, working, and simply going around in the environment (e.g., cognitive vision for human activity interpretation). In these contexts, fundamental topics and research questions in spatial representation and reasoning, commonsense reasoning, formal semantics & conceptual inference, and reasoning about action and change shall be presented.
Once the significance of space, events, actions, and change is established, I will present a general model of “Narrative Cognition” for computational cognitive systems wrt. the range of domains being addressed, i.e., vision, robotics, design, geography, where a general “narrative structure” encompassing spatio-temporal dynamics clearly emerges. Driven by the significance of the “narrative form”, and by the ubiquity of spatio-temporal narrativisation in a range of dynamic phenomena, I propose to ontologically and formally characterise a “computational model of narratives of space, actions, events, and change” as a basis for visual and spatio-temporal thinking in a wide-range of application areas.
I conclude by positioning that at a methodological level, the proposed computational model of narrative cognition presents a compelling approach toward integrating cognitively motivated theories of thinking and reasoning (e.g., theory of mental models) with formal methods in artificial intelligence for knowledge representation and reasoning & learning (e.g., about dynamic spatio-temporal phenomena). |
25.11.2014 |
Institute meeting |
|
|
02.12.2014 |
Cognitive usability factors of spatial interfaces and interfaces in space |
Sven Bertel |
Universität Weimar |
|
For various spatial tasks in human-computer interaction, measures of performance or emotion can be improved by adapting the user interface to a user’s individual cognitive profile. Such tasks can be found, for example, with human-computer collaboration in spatial reasoning or problem solving, information visualization, or gaming. In this talk, I will present work on past and present projects at the Usability Research Lab in Weimar, in which usability factors important for selected spatial tasks were investigated. These tasks involve mental and physical object rotation and the finding of map locations with the help of different graphical pointers. A secondary focus will be on challenges of designing effective adaptive systems based on knowledge gained about interindividual differences in spatial problem solving.
Information about the Usability Research Lab at Bauhaus-Universität Weimar can be found at www.uni-weimar.de/usability. |
09.12.2014 |
Wayfinding Through Orientation |
Stefan Münzer
Angela Schwering
|
University of Mannheim
Universität Münster
|
|
Wayfinding involves memory for routes, general orientation in an environment and the acquisition of survey knowledge. To date, mobile car and pedestrian navigation assistance systems offer guidance by instructing the navigator turn-by-turn based on a metric map. Those systems do not provide comprehensive spatial overview and orientation information. Therefore, they do not support the formation of a cognitive map. This research project develops a new paradigm for assisted navigation termed Wayfinding Through Orientation. This paradigm supports the navigator's wayfinding activities through providing orientation information that is cognitively appropriate by resembling attributes of a cognitive map. |
16.12.2014 |
From Maps to Geomedial Action Spaces – Theoretical and Methodological Considerations of Contemporary Mapping Practices |
Pablo Abend |
|
|
The increased networked digital distribution of geographic information for consumer-orientated cartographic representations leads to substantial changes in how people individually and collaboratively experience and produce space and place. Maps change from offering static and non-interactive frames of geographic reference for the production of space and place to support a veritable infinity of interactive and map-based activities associated with geomedia. The talk addresses significant questions in geography and cognate fields about the future roles of maps in the information age drawing on a qualitative analysis on the usage of the most widely used type of consumer software Google Earth. The usage practice termed geobrowsing, in particular, facilitates new modes of interaction that move from logocentric engagements with static maps to egocentric dynamic interactions with code-based elements of geomedial action spaces. The talk focuses on this ongoing metamorphosis seen in geobrowsing, the media-based flexible production of geographic knowledge through interactive maps. Drawing on work in media studies influenced by the so-called spatial turn – the engagement with geography-related questions in the social sciences and humanities –, a theoretical framework is introduced built on the concept of geomedial action spaces to understand the changing roles of information age maps as imagined materialist spaces for the experience and production of space. |
06.01.2015 |
kein GI-Forum |
|
|
13.01.2015 |
Social Interaction over a distance with multimodal displays |
Susanne Boll |
University of Oldenburg |
|
Social interaction and the feeling of relatedness are decisively set up by us communicating with fellow humans. Today, explicit communication, like direct conversation, the exchange via email or social networks, is already supported by information and communication technology. Nevertheless, at least as important is the implied, non- verbal communication, f. e. the incidental perception of the emotional state, a performed activity, the environment or the presence of other people. So far and unfortunately, there is no technical support whatsoever for implied communication between spatially separated persons. This is disadvantageous to people who are terminally ill, for social interaction and communication mean a lot regarding quality of life.
In the project CONTACT, which is funded by the BMBF, aims to develop an assistive system to stimulate social interaction of palliative patients who live alone in order to promote the implied communication and the exchange of non-verbal situations (emotions and activities) between patients, relatives and friends. In addition, human-machine-interactions are formed multimodal, affectively and all pervasively for the home environment which adapt to the person and the situation. For example, a lamp might inform relatives about the patient´s emotional state by changing its color or brightness. Furthermore, picture frames, which are equipped with sound, may indicate that the people on the pictures are thinking about the patient, and the coffeemaker or toaster being used shows that breakfast is being prepared. Accordingly, these devices will light up and, thus, generate a feeling of emotional closeness.
In this talk I will present requirements, design studies and prototypes for implicit social interaction using ubiquituous multimodal displays.
|
20.01.2015 |
Institute meeting |
|
|
27.01.2015 |
fällt leider aus |
Michael Gould |
Universitat Jaume I (UJI), Educational Manager of ESRI |
03.02.2015 |
|
Carl Schultz und Jakub Krukar |
Universität Münster |