Similarity-based analysis for large networks of ultra-low resolution sensors

Wren, Christopher R.; Minnen, David; Rao, Srinivasa G.

doi:10.1016/j.patcog.2006.04.009

Cited by 23 publications

(12 citation statements)

References 13 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…In either case, executing those policies is likely to be detrimental to the performance of the system. It has been shown that one can reliably recover rough topology of these kinds of sensor networks [25,14], so discarding event-pairs from non-adjacent sensors, for example, should be quite possible, and should result in a significant improvement in performance. We discuss such a scheme in Sect.…”

Section: Discussionmentioning

confidence: 99%

Functional calibration for pan-tilt-zoom cameras in hybrid sensor networks

2006

Self Cite

View full text Add to dashboard Cite

Wide-area context awareness is a crucial enabling technology for next generation smart buildings and surveillance systems. It is not practical to gather this context awareness by covering the entire building with cameras. However, significant gaps in coverage caused by installing cameras in a sparse way can make it very difficult to infer the missing information. As a solution we advocate a class of hybrid perceptual systems that build a comprehensive model of activity in a large space, such as a building, by merging contextual information from a dense network of ultra-lightweight sensor nodes and video from a sparse network of cameras. In this paper we explore the task of automatically recovering the relative geometry between a pan-tilt-zoom camera and a network of one-bit motion detectors. We present results both for the recovery of geometry alone and also for the recovery of geometry jointly with simple activity models. Because we do not believe a metric calibration is necessary, or even entirely useful, for this task, we formulate and pursue the novel goal we term functional calibration. Functional calibration is a blending of geometry estimation and simple behavioral model discovery. Accordingly, results are evaluated by measuring the ability of the system to automatically foveate targets in a large, non-convex space, rather than by measuring, for example, pixel reconstruction error.

show abstract

Section: Discussionmentioning

confidence: 99%

Functional calibration for pan-tilt-zoom cameras in hybrid sensor networks

2006

Self Cite

View full text Add to dashboard Cite

show abstract

“…Pattern recognition methods are adapted to fi nd spatio-temporal patterns in multiple streams of sensor data for automatic analysis of human behaviours and habits in these settings. These methods include search for recurrent event patterns (Magnusson 2000 ;Tavenard et al 2007 ) , clustering time series generated by lowresolution sensors using Markov models (Wren et al 2006 ) , using compression algorithms for extracting patterns (Cook 2005 ) , and eigen-analysis of behaviours (Eagle and Pentland 2006 ) .…”

Section: Behavioural Biometrics and Applicationsmentioning

confidence: 99%

Behavioural Biometrics and Human Identity

Schouten

Salah

Kranenburg

2012

The International Library of Ethics, Law and Technology

View full text Add to dashboard Cite

“…Probabilistic frameworks are aimed at recognizing complex contexts using data originating from a limited amount of highly expressive information sources, for example by using Bayesian state estimation techniques [1,26,27,[36][37][38]. Obvious drawbacks associated with these approaches are due to the possibly high error probability rates characterizing both measurement and data association processes, and to the difficulty of providing numerical data with semantic labelling [13].…”

Section: Related Workmentioning

confidence: 99%

A cognitive model for recognizing human behaviours in smart homes

2010

View full text Add to dashboard Cite

This paper describes strategies for monitoring human behaviour within a smart home and, in a broader perspective, for context–assessment in cognitive systems. The proposed framework, which is inspired by a cognitive theory called Computational Functionalism, is aimed at integrating ontology and logic based approaches to context representation\ud and recognition. Two are the assumptions underlying the model: (i) the availability of an ontology (i.e., a “concept–role” representation of what is relevant in a given domain); (ii) a\ud simple inference scheme (i.e., subsumption between descriptions of elements within the ontology). The context model is formally defined adopting a structural approach that describes contexts and situations as hierarchical structures grounded with respect to the ontology. Examples are presented to discuss the proposed model

show abstract

Similarity-based analysis for large networks of ultra-low resolution sensors

Cited by 23 publications

References 13 publications

Functional calibration for pan-tilt-zoom cameras in hybrid sensor networks

Functional calibration for pan-tilt-zoom cameras in hybrid sensor networks

Behavioural Biometrics and Human Identity

A cognitive model for recognizing human behaviours in smart homes

Contact Info

Product

Resources

About