Compressed-domain Fall Incident Detection for Intelligent Homecare
Journal of VLSI Signal Processing Systems
Video Monitoring of Vulnerable People in Home Environment
ICOST '08 Proceedings of the 6th international conference on Smart Homes and Health Telematics
Modelling Scenes Using the Activity within Them
Proceedings of the international conference on Spatial Cognition VI: Learning, Reasoning, and Talking about Space
Automated detection of unusual events on stairs
Image and Vision Computing
Navigational strategies in behaviour modelling
Artificial Intelligence
Concept and Design of a Video Monitoring System for Activity Recognition and Fall Detection
ICOST '09 Proceedings of the 7th International Conference on Smart Homes and Health Telematics: Ambient Assistive Health and Wellness Management in the Heart of the City
ANGELAH: a framework for assisting elders at home
IEEE Journal on Selected Areas in Communications - Special issue on wireless and pervasive communications for healthcare
Automatic detection of human fall in video
PReMI'07 Proceedings of the 2nd international conference on Pattern recognition and machine intelligence
Scene modelling and classification using learned spatial relations
COSIT'09 Proceedings of the 9th international conference on Spatial information theory
Goal-based trajectory analysis for unusual behaviour detection in intelligent surveillance
Image and Vision Computing
Building semantic scene models from unconstrained video
Computer Vision and Image Understanding
Robust fall detection by combining 3d data and fuzzy logic
ACCV'12 Proceedings of the 11th international conference on Computer Vision - Volume 2
Hi-index | 0.00 |
Interpretation of human activity and the detection of associated events are eased if appropriate models of context are available. A method is presented for automatically learning a context-specific spatial model in terms of semantic regions, specifically inactivity zones and entry zones. Maximium a posteriori estimation of Gaussian mixtures is used in conjunction with minumum description length for selection of the number of mixture components. Learning is performed using expectation-maximisation algorithms to maximise penalised likelihood functions that incorporate prior knowledge of the size and shape of the semantic regions. This encourages a one-to-one correspondence between the Gaussian mixture components and the regions. The resulting contextual model enables human-readable summaries of activity to be produced and unusual inactivity to be detected. Results are presented using overhead camera sequences tracked using a particle filter. The method is developed and described within the context of supportive home environments which have as their aim the extension of independent, quality living for older people.