Layered representations for human activity recognition

Author

Oliver, Nuria ; Horvitz, Eric ; Garg, Ashutosh

Author_Institution

Adaptive Syst. & Interaction, Microsoft Res., Redmond, WA, USA

fYear

2002

fDate

2002

Firstpage

3

Lastpage

8

Abstract

We present the use of layered probabilistic representations using hidden Markov models for performing sensing, learning, and inference at multiple levels of temporal granularity We describe the use of representation in a system that diagnoses states of a user´s activity based on real-time streams of evidence from video, acoustic, and computer interactions. We review the representation, present an implementation, and report on experiments with the layered representation in an office-awareness application.

Keywords

acoustic signal processing; hidden Markov models; inference mechanisms; learning (artificial intelligence); office automation; real-time systems; sensor fusion; user interfaces; video signal processing; acoustic interactions; computer interactions; hidden Markov models; human activity recognition; inference; layered probabilistic representations; learning; office-awareness application; real-time evidence streams; sensing; temporal granularity; video interactions; Adaptive systems; Application software; Context; Hidden Markov models; Humans; Machinery; Real time systems; Streaming media; Surveillance; Traffic control;

fLanguage

English

Publisher

ieee

Conference_Titel

Multimodal Interfaces, 2002. Proceedings. Fourth IEEE International Conference on

Print_ISBN

0-7695-1834-6

Type

conf

DOI

10.1109/ICMI.2002.1166960

Filename

1166960