63,976 Hits in 12.5 sec

Classification of Local Field Potentials using Gaussian Sequence Model [article]

Taposh Banerjee, John Choi, Bijan Pesaran, Demba Ba, Vahid Tarokh
2017 arXiv   pre-print
A problem of classification of local field potentials (LFPs), recorded from the prefrontal cortex of a macaque monkey, is considered.  ...  The theory of Gaussian sequence models allows us to represent minimax estimators as finite dimensional objects.  ...  CONCLUSIONS AND FUTURE WORK We proposed a new framework for robust classification of local field potentials and used it to obtain Pinsker's and blockwise James-Stein classifiers.  ... 
arXiv:1710.01821v4 fatcat:2ifesqe5ezaxfkjop6phtszbqa

Spatiotemporal Algorithm For Joint Video Segmentation And Foreground Detection

Sevket Derin Babacan, Thrasyvoulos Pappas
2006 Zenodo  
Publication in the conference proceedings of EUSIPCO, Florence, Italy, 2006  ...  The mixture of Gaussians in ( 8 ) is used to model the background pixel intensities.  ...  We obtain an initial estimate of x of the first frame using the Markov random field model discussed in Section 2.1.  ... 
doi:10.5281/zenodo.52674 fatcat:acmxmonf3javjgz5aeaiz64p5a

An ICA Mixture Hidden Conditional Random Field Model for Video Event Classification

Xiaofeng Wang, Xiao-Ping Zhang
2013 IEEE transactions on circuits and systems for video technology (Print)  
In addition, according to the non-Gaussian property of video event features, a new feature function using the likelihoods of ICA mixture components is proposed for local observation to further enhance  ...  The discriminative power of the HCRF and representation power of the ICA mixture for non-Gaussian distributions are combined in the new model.  ...  A New Hidden Conditional Random Field Model for Video Event Classification A.  ... 
doi:10.1109/tcsvt.2012.2203195 fatcat:7yvmwwzalneslonfyn6oudocpa

ICA mixture hidden conditional random field model for sports event classification

Xiaofeng Wang, Xiao-Ping Zhang
2009 2009 IEEE 12th International Conference on Computer Vision Workshops, ICCV Workshops  
According to the non-Gaussian property of sports event features, a new feature function using the likelihood of ICA mixture component is proposed to further enhance the HCRF model.  ...  The discriminant power of HCRF and representation power of ICA mixture for non-Gaussian distribution are combined. The new model is applied to challenging bowling and golf event classification.  ...  Unlike the Gaussian assumption in other works [4] [12], we use a non Gaussian model as local feature function for observations.  ... 
doi:10.1109/iccvw.2009.5457653 dblp:conf/iccvw/WangZ09 fatcat:u4kkfm5twbc7dg3yz45v37vqta

A probabilistic patch based image representation using Conditional Random Field model for image classification [article]

Fariborz Taherkhani
2016 arXiv   pre-print
Second, the sequence of these ordered patches is modeled as a probabilistic feature vector by CRF to model spatial relationship of these local properties.  ...  In this paper we proposed an ordered patch based method using Conditional Random Field (CRF) in order to encode local properties and their spatial relationship in images to address texture classification  ...  To include spatial relationships of these local appearance for image representation we use conditional random field (CRF) model.  ... 
arXiv:1607.06797v2 fatcat:vtqjfaobbraqba3gccbhgh4dca

Motion compensated color video classification using Markov Random Fields [chapter]

Zoltan Kato, Ting-Chuen Pong, John Chung-Mong Lee
1997 Lecture Notes in Computer Science  
This paper deals with the classification of color video sequences using Markov Random Fields (MRF) taking into account motion information.  ...  The sequence is regarded as a stack of frames and both intra-and inter-frame cliques are defined in the label field.  ...  The classification model is defined in a Markovian framework and uses a first order potential derived from a three-variate Gaussian distribution in order to tie the final classification to the observed  ... 
doi:10.1007/3-540-63930-6_189 fatcat:nzn34udy55cczbpq2uw75mifge

Markov random fields for abnormal behavior detection on highways

P. L. M. Bouttefroy, A. Beghdadi, A. Bouzerdoum, S. L. Phung
2010 2010 2nd European Workshop on Visual Information Processing (EUVIP)  
Contrary to traditional methods, the proposed technique models the local density of object feature vector, therefore leading to simple and elegant criterion for behavior classification.  ...  We develop a Gaussian Markov random field mixture catering for multi-modal density and integrating the neighborhood behavior into a local estimate.  ...  Gaussian Markov Random Field Mixture The Gaussian Markov random field provides a unimodal estimate of the local behavior through the clique and spatial potentials.  ... 
doi:10.1109/euvip.2010.5699125 dblp:conf/euvip/BouttefroyBBP10 fatcat:rznp3u275zcd7luj4il7ctmrea

Conditional Random Field for Natural Scene Categorization

Y. Wang, s. Gong
2007 Procedings of the British Machine Vision Conference 2007  
Conditional random field (CRF) has been widely used for sequence labeling and segmentation. However, CRF does not offer a straightforward approach to classify whole sequences.  ...  On the other hand, hidden conditional random field (HCRF) has been proposed for whole sequences classification by viewing the segment labels as hidden variables.  ...  A good candidate for modeling a set of ordered local patches is the conditional random field (CRF) [5] .  ... 
doi:10.5244/c.21.59 dblp:conf/bmvc/WangG07 fatcat:7aolirbjkfabjfnmmtj2lux3hq

Support Vector Random Fields for Spatial Classification [chapter]

Chi-Hoon Lee, Russell Greiner, Mark Schmidt
2005 Lecture Notes in Computer Science  
In this paper we propose Support Vector Random Fields (SVRFs), an extension of Support Vector Machines (SVMs) that explicitly models spatial correlations in multi-dimensional data.  ...  We also propose improvements to computing posterior probability distributions from SVMs, and present a local-consistency potential measure that encourages spatial continuity.  ...  Greiner is supported by the National Science and Engineering Research Council of Canada (NSERC) and the Alberta Ingenuity Centre for Machine Learning (AICML). C.H.  ... 
doi:10.1007/11564126_16 fatcat:wvoildwjtrguvepsmww26jrt64

Gaussian Process Based Motion Pattern Recognition with Sequential Local Models

Mattias Tiger, Fredrik Heintz
2018 2018 IEEE Intelligent Vehicles Symposium (IV)  
We investigate the impact of varying local model overlap and the length of the observed trajectory trace on the classification quality.  ...  In this paper we consider the problem of motion pattern recognition in the setting of sequential local motion pattern models.  ...  Also, the use of sequential local models makes a longer sequence more likely to overlap two neighboring sequential local models. The sequence length 1 is used in the larger experiment.  ... 
doi:10.1109/ivs.2018.8500676 dblp:conf/ivs/TigerH18 fatcat:iirdvgd5qnbhrbysk72trezmiq

A Discriminative Framework for Action Recognition Using f-HOL Features

Samy Bakheet, Ayoub Al-Hamadi
2016 Information  
The extracted features are then fed into a discriminative conditional model based on Latent-Dynamic Conditional random fields (LDCRFs) to learn to recognize actions from video frames.  ...  Inspired by the overwhelming success of Histogram of Oriented Gradients (HOG) features in many vision tasks, in this paper, we present an innovative compact feature descriptor called fuzzy Histogram of  ...  We would also like to thank anonymous reviewers for their thorough reading of the manuscript and their insightful comments and suggestions.  ... 
doi:10.3390/info7040068 fatcat:twbkniu4ivbdzfeqcpzisl4dma

Locally Smoothed Neural Networks [article]

Liang Pang, Yanyan Lan, Jun Xu, Jiafeng Guo, Xueqi Cheng
2017 arXiv   pre-print
Specifically, a multi-variate Gaussian function is utilized to generate the smoother, for modeling the location relations among different local receptive fields.  ...  is for determining the importance and relations of different local receptive fields.  ...  In order to capture the location relations among different local receptive fields, i.e. neighbouring fields should be similar, a multi-variate Gaussian function is used to generate the smoother.  ... 
arXiv:1711.08132v1 fatcat:cfofru4zajfqzg4qvuoinopzou

Automatic Detection of Adverse Weather Conditions in Traffic Scenes

Andrea Lagorio, Enrico Grosso, Massimo Tistarelli
2008 2008 IEEE Fifth International Conference on Advanced Video and Signal Based Surveillance  
The developed system uses a statistical framework based on the mixture of Gaussians to identify changes both in the spatial and temporal frequencies which characterize specific meteorological events.  ...  In this scenario, the analysis of weather conditions is considered to signal particular and potentially dangerous situations like the presence of snow, fog, or heavy rain.  ...  Among them the most commonly used method is based on the mixture of Gaussian (MOG) model [9, 10] .  ... 
doi:10.1109/avss.2008.50 dblp:conf/avss/LagorioGT08 fatcat:qxascaymebfxvlt6doodsq5bqi

Fluid Dynamic Models for Bhattacharyya-Based Discriminant Analysis

Yung-Kyun Noh, Jihun Hamm, Frank Chongwoo Park, Byoung-Tak Zhang, Daniel D. Lee
2018 IEEE Transactions on Pattern Analysis and Machine Intelligence  
We consider the problem of reducing the dimensionality of labeled data for classification.  ...  We derive the forces arising in the fluids from information theoretic potential functions, and consider appropriate low rank constraints on the resulting acceleration and velocity flow fields.  ...  We next use this equation to derive the resulting force field corresponding to a given potential function.  ... 
doi:10.1109/tpami.2017.2666148 pmid:28186879 fatcat:ifurm7zlbbe7jblnna4bsvduty

Regularization, adaptation, and non-independent features improve hidden conditional random fields for phone classification

Yun-Hsuan Sung, Constantinos Boulis, Christopher Manning, Dan Jurafsky
2007 2007 IEEE Workshop on Automatic Speech Recognition & Understanding (ASRU)  
We show a number of improvements in the use of Hidden Conditional Random Fields (HCRFs) for phone classification on the TIMIT and Switchboard corpora.  ...  We then show that HCRFs are able to make use of non-independent features in phone classification, at least with small numbers of mixture components, while HMMs degrade due to their strong independence  ...  Conditional Random Fields (CRFs) [2] are another widelyused sequence labeling model that are attractive as a potential replacement for HMMs.  ... 
doi:10.1109/asru.2007.4430136 dblp:conf/asru/SungBMJ07 fatcat:mx4rs4tv5jcinpdw62wdm3jxpy
« Previous Showing results 1 — 15 out of 63,976 results