Location-Based Activity Recognition using Relational Markov Networks

By Lin Liao, Dieter Fox, Henry A. Kautz. International Joint Conference on Artificial Intelligence, 2005

378 citations recorded by the source corpus.

View Paper: https://www.semanticscholar.org/paper/d3ca30648ac4dcf1dd62d11d520ccd5f18b0c051

How researchers see this paper

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

Citations in the corpus

Learning Setting-Generalized Activity Models for Smart Spaces (2010, methodology): “5,6,7,8 In our approach we initially test three models: a naïve Bayes classifier (NBC), a hidden Markov model (HMM), and a conditional random field (CRF) model.”
Assessing the Quality of Activities in a Smart Environment (2009, background): “Researchers have investigated both the use of attribute/ value-based learners [11] and temporal models [12] for activity recognition.”
Extracting Places and Activities from GPS Traces Using Hierarchical Conditional Random Fields (2007, methodology): “Parameter sharing can be modeled conveniently using probabilistic relational models such as relational Markov networks (Taskar et al 2002; Liao et al. 2005c).”
Egocentric recognition of handled objects: Benchmark and analysis (2009, background): “In the ubiquitous and pervasive computing communities, inferring the activities of a user is an important and popular problem and has been extensively studied, often using alternative sensors such as accelerometers [14, 12] or GPS [16].”
COSAR: hybrid reasoning for context-aware activity recognition (2011, background): “Spatiotemporal traces are used in [18] to derive high-level activities such as shopping or dining out.”
PEIR, the personal environmental impact report, as a platform for participatory sensing systems research (2009, methodology): “The work of [16, 17] uses GPS data with external knowledge about bus routes and bus stops to infer and predict a user’s transportation mode such as walking, driving, or taking a bus, and in [18, 19] models of a user’s activities and places from traces of GPS data and locations of restaurant, stores,”
Categorizing object-action relations from semantic scene graphs (2010, methodology): “Similar to this study, Liao et al. (2005) provided an approach to perform location-based activity recognition by using Relational Markov Networks [13].”
Wireless Sensor Networks for Healthcare (2010, background): “While wireless healthcare systems using various wireless technologies exist, this work focuses on systems based on low-power wireless platforms for physiological and motion monitoring studies, and smartphone-based large-scale studies.”
A Survey on Ambient-Assisted Living Tools for Older Adults (2013, background): “the sequential nature of data, including Markov chains [72], Dynamic Bayesian network [73]–[75], hidden Markov model (HMM) [71], and conditional random fields (CRFs) [76], [77].”
Learning the semantics of object–action relations by observation (2011, background): “…approaches have been presented for vision-based recognition of manipulations (which is the focus of the work presented in this paper), vision-based recognition of human-motion patterns, and non-visual recognition of other types of activities (Hongeng 2004; Liao et al. 2005; Modayil et al. 2008).”
Semantic trajectories modeling and analysis (2013, background): “For example, Liao et al. [2005] use a Gaussian mixture model based on the speed, which is classified into three speed ranges: walking, high speed, and low speed. Therefore, a new episode is created whenever a switch in the speed range is detected. A similar approach is developed in Zheng et al. [201”
Inference Attacks on Location Tracks (2007, background): “Liao et al.[21] used this algorithm to find a user’s frequent destinations for higher-level machine learning about a user’s habits.”
Conditional random fields for activity recognition (2007, background): “As we discuss in chapter 6, relational Markov networks have been applied to activity recognition [64].”
Recruitment Framework for Participatory Sensing Data Collections (2010, background): “Also, certain systems use map matching and reverse geo-coding to add additional contextual information (semantic meaning) to the clusters [15].”
Activity Recognition on Streaming Sensor Data (2012, background): “These sensors are adept in performing location based activity recognition in indoor environments just as GPS is used for outdoor environments [21].”
Wireless Architectures for Heterogeneous Sensing in Smart Home Applications: Concepts and Real Implementation (2013, methodology): “Currently used methods are mostly based on machine-learning models, such as Bayes classifiers [54], [61], Markov models [62], [63], artificial neural networks [64], [65], and support vector machines (SVMs) [66], [67], thanks to their ability to create context models from measured data and adapt…”
Unsupervised Activity Recognition Using Automatically Mined Common Sense (2005, background): “Liao, Fox, & Kautz (2005) have shown that learning models from different individuals can improve activity recognition for a new individual, and we expect accuracy to improve even more for the case where all of the training data comes from a single person.”
Keeping the Resident in the Loop: Adapting the Smart Home to the User (2009, methodology): “To achieve intelligent behaviors in smart-home applications, various computational intelligence techniques have been proposed to support the creation of smart homes, including neural networks [6], fuzzy logic [7], hierarchical task network planning [8], hidden Markov models [9], and Bayesian network”