October 9-12, Anchorage, USA
This workshop will provide a venue to discuss robust machine learning techniques for human activity recognition using body worn sensors, as well as the need for proper benchmarking datasets and tools
Human activity recognition can be used to devise assistants that provide proactive support by exploiting the knowledge of the user’s context, determined from sensors located on-body. The design and development of these systems pose important challenges to the machine learning community as they typically involve high-dimensional, multimodal streams of data characterized by a large variability; where data portions may be missing or labels can be unreliable.
Notwithstanding the large amount of research endeavors aimed at tackling these issues, the comparison of different approaches is often not possible due to the lack of common benchmarking tools and datasets that allow for replicable and fair testing procedures across several research groups. The aim of this workshop is to discuss and compare different methods for robust activity recognition, as well as putting forward the need for common resources for such comparison. To promote such comparison, the workshop is associated to an activity recognition challenge where contributed methods will be evaluated on a public benchmark database of daily activities recorded using a multimodal network of on-body sensors.
Prof. Nobuo Kawaguchi, Nagoya University, Japan
Dr. Kamil Kloch / Prof. Dr. Paul Lukowicz, Universität Passau, Germany
Dr. Kristof Van Laerhoven, Technische Universität Darmstadt, Germany
Dr. Thomas Ploetz , Newcastle University, UK; Georgia Tech, USA
June 28, 2011: Regular papers submission deadline
July 1, 2011: Acceptance/rejection notification
July 5, 2011: Camera ready version due
August 31, 2011: 'Last-minute-results' abstracts submission deadline
September 12, 2011: 'Last-minute-results' acceptance/rejection notification
Oct 9, 2011: Workshop at the SMC Conference
Dr. Ricardo Chavarriaga, EPFL, Lausanne, Switzerland
Dr. Daniel Roggen, ETHZ, Zürich, Switzerland
Prof. Dr. Alois Ferscha, JKU, Linz, Austria
We call for contributions addressing specific problems in activity recognition from body-worn sensors. Each paper should be concise with sufficient detail and references to allow critical review and formatted according to the conference guidelines (max 4 pages). Papers will be reviewed by two referees for technical merit and content. Accepted papers will appear in the conference proceedings only if one of the authors is registered for the conference and presents the paper at the workshop.
Participants will be able to present and discuss their most recent developments during a last-minute results session within the workshop. Contributors should submit a 2-page abstract that will be reviewed by the organizing committe. Depending on the availability of time, outstanding contributions will be considered for oral presentation. Please notice that the abstracts will be disrtibuted to participants but will not be included in the conference proceedings. Submissions should be send to email address: firstname.lastname@example.org.
In order to allow comparison of about the benefits/drawbacks of different approaches under the same testing conditions we strongly encourage the submission of papers that address the problems proposed in the associated activity recognition challenge.
Do not hesitate to contact the consortium.
We develop opportunistic activity recognition systems: goal-oriented sensor assemblies spontaneously arise and self-organize to achieve a common activity and context recognition. We develop algorithms and architectures underlying context recognition in opportunistic systems.
Subscribe to our newsletter for regular project updates.