Layered representations for learning and inferring office activity from multiple sensory channels

By Nuria Oliver, Ashutosh Garg, Eric Horvitz. 2004

365 citations recorded by the source corpus.

Subfields: Artificial Intelligence, Computer Vision and Pattern Recognition, Signal Processing

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Artificial Intelligence

In Computer Vision and Pattern Recognition

In Signal Processing

Citations in the corpus

Deep Learning: Methods and Applications (2014, methodology): “Related models with greater technical depth and mathematical treatment can be found in [116] for HHMM and [271] for Layered HMM.”
A tutorial survey of architectures, algorithms, and applications for deep learning (2014, methodology): “Relatedmodels with greater technical depth andmathematical treatment can be found in [83] for HHMM and [84] for Layered HMM.”
Searching Video for Complex Activities with Finite State Models (2007, methodology): “Methodsinclude: coupled HMM’s ([10]; to classify T’ai Chi moves); layered HMM’s ([ 22 ]; to represent office activity); hierachies ([21]; to recognize everyday gesture); HMM’s with a global free parameter ([29]; to model gestures); and entropic HMM’s ([9]; for video puppetry).”
A Scalable Approach to Activity Recognition based on Object Use (2007, methodology): “Temporal constraints on actions are addressed either through the use of probabilistic models such as HMM’s [4, 16] or SCFG’s [9], or through explicit temporal correlation methods [21, 26].”
Automatic analysis of multimodal group actions in meetings (2005, methodology): “Visual recognition of american sign language using HMMs.”
Sensor-Based Activity Recognition (2012, methodology): “Coupled Hidden Markov Models (CHMMs) have two Markovian chains, each modeling a different stream of data, with a coupling between them to model their interdependence .”
Expandable Data-Driven Graphical Modeling of Human Actions Based on Salient Postures (2008, methodology): “Ahmad and Lee [40] built multiple HMMs for each action, each HMM being for the action observed from a particular viewpoint.”
Human Action Recognition by Semilatent Topic Models (2009, positive): “Feng and Perona [32] build HMM models from vector-quantized image shapes called “movelets.” Olivera et al. [ 33 ] use a layered HMM to represent office activities.”
EnsembleMatrix: interactive visualization to support machine learning with multiple classifiers (2009, neutral): “On the other spectrum lie critic-driven approaches, such as layered HMMs [26], where the goal is to “learn” a good combination scheme using a hierarchy of classifiers.”
Classifying actions based on histogram of oriented velocity vectors (2015, background): “The results of their implementation have shown that HCRFs outperformed both CRFs and HMMs.”
Social signal processing: Survey of an emerging domain (2009, methodology): “the combination of the information extracted from different modalities is performed with algorithms Dynamic Bayesian Networks [126] and layered Hidden Markov Models [130].”
Modelling Crowd Scenes for Event Detection (2006, background): “Moreover it is difficulty to track individuals in the crowd using state of the art tracking algorithms.”
Context-Aware Computing, Learning, and Big Data in Internet of Things: A Survey (2018, methodology): “[169] used HMM to determine office activities from multiple sensory channels.”
Learning Motion Patterns of People for Compliant Robot Motion (2005, background): “Oliver, Horvitz, and Garg (2002) use data obtained from various sensors as input to a layered HMM and infer the state of a user’s activity.”
Automatic nonverbal analysis of social interaction in small groups: A review (2009, methodology): “First, HMMs can be challenged by a large number of parameters, and by the risk of overfitting when learned from limited data [95].”
A Systematic Review of Hidden Markov Models and Their Applications (2020, methodology): “Each layer is connected to the next layer through inferential results [115].”
Parsing Videos of Actions with Segmental Grammars (2014, methodology): “HMMs are recursive HMMs where each hi den state is its own sequential probab listic model.”