Hier finden Sie wissenschaftliche Publikationen aus den Fraunhofer-Instituten.

Activity recognition from sparsely labeled data using multi-instance learning

: Stikic, M.; Schiele, B.


Choudhury, T.:
Location and context awareness. 4th international symposium, LoCA 2009 : Tokyo, Japan, May 7 - 8, 2009
Berlin: Springer, 2009 (Lecture Notes in Computer Science 5561)
ISBN: 3-642-01720-7 (print)
ISBN: 978-3-642-01721-6
ISSN: 0302-9743
International Symposium on Location and Context Awareness (LoCA) <4, 2009, Tokyo>
Fraunhofer IGD ()

Activity recognition has attracted increasing attention in recent years due to its potential to enable a number of compelling contextaware applications. As most approaches rely on supervised learning methods, obtaining substantial amounts of labeled data is often an important bottle-neck for these approaches. In this paper, we present and explore a novel method for activity recognition from sparsely labeled data. The method is based on multi-instance learning allowing to significantly reduce the required level of supervision. In particular we propose several novel extensions of multi-instance learning to support different annotation strategies. The validity of the approach is demonstrated on two public datasets for three different labeling scenarios.