Hidden Conditional Random Fields for Gesture Recognition

By Sy Bor Wang, Ariadna Quattoni, Louis-philippe Morency, David Demirdjian, Trevor Darrell. 2006

562 citations recorded by the source corpus.

Subfields: Computer Vision and Pattern Recognition, Developmental and Educational Psychology, Human-Computer Interaction

How researchers see this paper

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

In Computer Vision and Pattern Recognition

In Developmental and Educational Psychology

In Human-Computer Interaction

Citations in the corpus

One-shot-learning gesture recognition using HOG-HOF features (2013, methodology): “The method improves overall results of our methods (Table 1), and small improvement was confirmed by Wu et al. (2012)—2% and Wan et al.”
Continuous body and hand gesture recognition for natural human-computer interaction (2012, background): “Many other variants of HCRFs have been introduced since then [Sutton et al. 2004; Gunawardana et al. 2005; Wang et al. 2006; Morency et al. 2007], but most of these were tested on single-signal pattern recognition tasks (e.”
Human Action Recognition by Semilatent Topic Models (2009, unknown): “Although generative models (e.g., HMM) were popular in sequence modeling, recently people have also been applying discriminative models for modeling temporal information, e.g., conditional random fields (CRF) in Sminchisescu et al. [36] and hidden conditional random fields (HCRF) in Wang et al. [37]”
Modeling Temporal Structure of Decomposable Motion Segments for Activity Classification (2010, background): “On the other hand, discriminative models of temporal context have also being applied for classification of simple motions in rather simplified environments [12,13,14,15].”
Hidden Conditional Random Fields (2007, methodology): “In the arm gesture domain, we used a data set of gestures defined for a virtual manipulation task as described in [23].”
Human Action Recognition and Prediction: A Survey (2018, background): “, 2007; Ikizler & Forsyth, 2007), conditional random fields (CRFs) (Sminchisescu et al., 2005; Wang et al., 2006;Wang&Suter, 2007;Morency et al., 2007) and structured support vector machine (SSVM) (Niebles et al.”
One-shot learning gesture recognition from RGB-D data using bag of features (2013, methodology): “Wang et al. (2006) then incorporated hidden state variables into the CRF model, namely, hidden conditional random field (HCRF).”
Gesture Modeling by Hanklet-Based Hidden Markov Model (2014, methodology): “Another model often used for action recognition is Conditional Random Fields (CRF)[16],[17],[18], which is a discriminative approach and has proven to be successful for recognition task.”
Canonical Correlation Analysis of Video Volume Tensors for Action Categorization and Detection (2009, methodology): “A novel method is proposed to measure video-to-video volume similarity by extending Canonical Correlation Analysis (CCA), a principled tool to inspect linear relations between two sets of vectors, to that of two multiway data arrays (or tensors).”
Conditional random fields for activity recognition (2007, methodology): “As newer models, conditional random fields have received less attention than HMMs or DBNs, but they are used for activity recognition as well as in related domains that require rich, continuous features such as image segmentation [54], object recognition [88], and gesture recognition [116].”
Interpreting multimodal referring expressions in real time (2016, background): “Many approaches to recognition use discriminative models [28, 23], which have been shown to be more accurate than their generative counterparts.”
Latent Hierarchical Model of Temporal Structure for Complex Activity Classification (2014, methodology): “Probabilistic graphical models were usually adopted to model the temporal structure of human activity or motion trajectories, such as Hidden Markov Models (HMMs) [5], [9], Hidden Conditional Random Fields (HCRFs) [10], [11], and Dynamic Bayesian Networks (DBNs) [4], [34].”
Predictive-Corrective Networks for Action Detection (2017, methodology): “These models generally rely on chain structured models that admit efficient inference, such as HMMs [33, 12] and CRFs [52].”
Learning latent structure for activity recognition (2014, background): “Hidden Markov Models (HMMs) [1], [2], Dynamic Bayesian Networks (DBNs) [9], linear-chain CRFs [10], loopy CRFs [3], Semi-Markov Models [4], and Hidden CRFs [11], [8].”
Hidden Markov Model on a unit hypersphere space for gesture trajectory recognition (2014, methodology): “, 2001), Hidden CRF (HCRF) (Wang et al., 2006), and Latent-Dynamic CRF (LDCRF) (Morency et al.”
Learning structural SVMs with latent variables (2009, methodology): “Recently, there has been some work on Conditional Random Fields (Wang et al., 2006) with latent variables.”