Activity Recognition from Sparsely Labeled Data Using Multi-Instance Learning

Activity recognition has attracted increasing attention in recent years due to its potential to enable a number of compelling context-aware 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.

[1]  Vladimir Vapnik,et al.  Statistical learning theory , 1998 .

[2]  Thomas Hofmann,et al.  Support Vector Machines for Multiple-Instance Learning , 2002, NIPS.

[3]  Uwe Hansmann,et al.  Pervasive Computing , 2003 .

[4]  Kent Larson,et al.  Activity Recognition in the Home Using Simple and Ubiquitous Sensors , 2004, Pervasive.

[5]  Zhi-Hua Zhou.  Multi-Instance Learning : A Survey , 2004 .

[6]  Ling Bao,et al.  Activity Recognition from User-Annotated Acceleration Data , 2004, Pervasive.

[7]  Brigham Anderson,et al.  Active learning for Hidden Markov Models: objective functions and algorithms , 2005, ICML.

[8]  Matthai Philipose,et al.  Unsupervised Activity Recognition Using Automatically Mined Common Sense , 2005, AAAI.

[9]  Svetha Venkatesh,et al.  Activity recognition and abnormality detection with the switching hidden semi-Markov model , 2005, 2005 IEEE Computer Society Conference on Computer Vision and Pattern Recognition (CVPR'05).

[10]  Jesse Hoey,et al.  Semi-supervised learning of a POMDP model of Patient-Caregiver Interactions , 2005 .

[11]  Michael L. Littman,et al.  Activity Recognition from Accelerometer Data , 2005, AAAI.

[12]  Paul Brna,et al.  User Modeling 2005, 10th International Conference, UM 2005, Edinburgh, Scotland, UK, July 24-29, 2005, Proceedings , 2005, User Modeling.

[13]  Eric Horvitz,et al.  A Comparison of HMMs and Dynamic Bayesian Networks for Recognizing Office Activities , 2005, User Modeling.

[14]  Emmanuel Munguia Tapia,et al.  Toward Scalable Activity Recognition for Sensor Networks , 2006, LoCA.

[15]  Alexander Zien,et al.  Semi-Supervised Learning , 2006 .

[16]  Irfan A. Essa,et al.  Discovering Characteristic Actions from On-Body Sensor Data , 2006, 2006 10th IEEE International Symposium on Wearable Computers.

[17]  Bernhard Schölkopf,et al.  Introduction to Semi-Supervised Learning , 2006, Semi-Supervised Learning.

[18]  Jeff A. Bilmes,et al.  Recognizing Activities and Spatial Context Using Wearable Sensors , 2006, UAI.

[19]  Gaetano Borriello,et al.  A Practical Approach to Recognizing Physical Activities , 2006, Pervasive.

[20]  Daniel P. Siewiorek,et al.  Activity recognition and monitoring using multiple sensors on different body positions , 2006, International Workshop on Wearable and Implantable Body Sensor Networks (BSN'06).

[21]  山崎 達也,et al.  Pervasive Computing for Quality of Life Enhancement, 5th International Conference On Smart Homes and Health Telematics, ICOST 2007, Nara, Japan, June 21-23, 2007, Proceedings , 2007, ICOST.

[22]  Bernt Schiele,et al.  Scalable Recognition of Daily Activities with Wearable Sensors , 2007, LoCA.

[23]  Matthai Philipose,et al.  Common Sense Based Joint Training of Human Activity Recognizers , 2007, IJCAI.

[24]  Donghai Guan,et al.  Activity Recognition Based on Semi-supervised Learning , 2007, 13th IEEE International Conference on Embedded and Real-Time Computing Systems and Applications (RTCSA 2007).

[25]  Maryam Mahdaviani,et al.  Fast and Scalable Training of Semi-Supervised CRFs with Application to Activity Recognition , 2007, NIPS.

[26]  James A. Landay,et al.  MyExperience: a system for in situ tracing and capturing of user feedback on mobile phones , 2007, MobiSys '07.

[27]  Juan Carlos Augusto,et al.  Distributed Vision-Based Accident Management for Assisted Living , 2007, ICOST.

[28]  Bernt Schiele,et al.  Location- and Context-Awareness, Third International Symposium, LoCA 2007, Oberpfaffenhofen, Germany, September 20-21, 2007, Proceedings , 2007, LoCA.

[29]  Jennifer Healey,et al.  A Long-Term Evaluation of Sensing Modalities for Activity Recognition , 2007, UbiComp.

[30]  Eric Horvitz,et al.  On Discarding, Caching, and Recalling Samples in Active Learning , 2007, UAI.

[31]  Philippe Golle,et al.  On using existing time-use study data for ubiquitous computing applications , 2008, UbiComp.

[32]  Bernt Schiele,et al.  Using rhythm awareness in long-term activity recognition , 2008, 2008 12th IEEE International Symposium on Wearable Computers.

[33]  Bernt Schiele,et al.  Discovery of activity patterns using topic models , 2008 .

[34]  Eric Horvitz,et al.  Experience sampling for building predictive user models: a comparative study , 2008, CHI.

[35]  Bernt Schiele,et al.  Exploring semi-supervised and active learning for activity recognition , 2008, 2008 12th IEEE International Symposium on Wearable Computers.

[36]  Paul Lukowicz,et al.  Using a complex multi-modal on-body sensor system for activity spotting , 2008, 2008 12th IEEE International Symposium on Wearable Computers.