Filters








1,330 Hits in 5.2 sec

Sequential Bayesian Nonparametric Multimodal Topic Models for Video Data Analysis

Jianfei XUE, Koji EGUCHI
2018 IEICE transactions on information and systems  
Dirichlet processes (Seq-CI-HDP) and sequential correspondence hierarchical Dirichlet processes (Seq-cHDP), to improve the multimodal data modeling mechanism via controlling the pivot assignments with  ...  Topic modeling as a well-known method is widely applied for not only text data mining but also multimedia data analysis such as video data analysis.  ...  Acknowledgments This work was supported in part by the Grant-in-Aid for Scientific Research (#15H02703) from JSPS, Japan.  ... 
doi:10.1587/transinf.2017dap0021 fatcat:oukkqsre2ncddgu6v6son5yh3i

Video Data Modeling Using Sequential Correspondence Hierarchical Dirichlet Processes

Jianfei XUE, Koji EGUCHI
2017 IEICE transactions on information and systems  
In this paper, we present a novel topic model named sequential correspondence hierarchical Dirichlet processes (Seq-cHDP) to learn the hidden structure within video data.  ...  machine learning, hierarchical Dirichlet processes, topic models Jianfei Xue is currently pursuing a Ph.D  ...  Acknowledgments This work was supported in part by the Grant-in-Aid for Scientific Research (#15H02703) from JSPS, Japan.  ... 
doi:10.1587/transinf.2016mup0007 fatcat:ige7axxttrantoikabolk4oavm

Fast unsupervised ego-action learning for first-person sports videos

Kris M. Kitani, Takahiro Okabe, Yoichi Sato, Akihiro Sugimoto
2011 CVPR 2011  
In our proposed framework we show that a stacked Dirichlet process mixture model can be used to automatically learn a motion histogram codebook and the set of ego-action categories.  ...  We address the novel task of discovering firstperson action categories (which we call ego-actions) which can be useful for such tasks as video indexing and retrieval.  ...  Comparative Evaluation We compare the top-level DPM with online inference (DPM-OL) against the DPM with variational inference (DPM-VI), latent Dirichlet allocation with variational inference and sequential  ... 
doi:10.1109/cvpr.2011.5995406 dblp:conf/cvpr/KitaniOSS11 fatcat:hf24ho4g7jfm3nxddavtf52s7i

[Invited papers] Supervised Nonparametric Multimodal Topic Models for Multi-class Video Classification

Jianfei Xue, Koji Eguchi
2019 ITE Transactions on Media Technology and Applications  
Nonparametric topic models such as hierarchical Dirichlet processes (HDP) have been attracting more and more attentions for multimedia data analysis.  ...  In this paper, we present a novel supervised sequential symmetric correspondence HDP (Sup-SSC-HDP) model for multi-class video classification, where the empirical topic frequencies learned from multimodal  ...  Acknowledgments This work was supported in part by the Grant-in-Aid for Scientific Research (#15H02703) from JSPS, Japan.  ... 
doi:10.3169/mta.7.80 fatcat:fe6wbherr5a6tc5m4ww2nvnlyu

An Improved Hierarchical Dirichlet Process-Hidden Markov Model and Its Application to Trajectory Modeling and Retrieval

Weiming Hu, Guodong Tian, Xi Li, Stephen Maybank
2013 International Journal of Computer Vision  
In this paper, we propose a hierarchical Bayesian model, an improved hierarchical Dirichlet process-hidden Markov model (iHDP-HMM), for visual document analysis.  ...  We apply the iHDP-HMM to simultaneously cluster motion trajectories and discover latent topics for trajectory words, based on the proposed method for constructing the trajectory word codebook.  ...  Hierarchical Dirichlet Process We briefly introduce the following document-word analysis techniques which are closely related to our work: the Dirichlet process model (DPM), the hierarchical Dirichlet  ... 
doi:10.1007/s11263-013-0638-8 fatcat:mn5abadpjraqvk2rz6krkcrnku

Infinite Hidden Markov Models and ISA Features for Unusual-Event Detection in Video

Iulian Pruteanu-Malinici, Lawrence Carin
2007 2007 IEEE International Conference on Image Processing  
A hierarchical Dirichlet process (HDP) framework is employed in the formulation of the iHMM.  ...  using "normal"/"typical" video data.  ...  A Dirichlet process (DP) model [9] is used to non-parametrically learn a GMM separately for each of the N data sets.  ... 
doi:10.1109/icip.2007.4379784 dblp:conf/icip/Pruteanu-MaliniciC07 fatcat:pmqs5bf5jvgidmc6ejyaexobzu

Dynamic Hierarchical Dirichlet Process for Abnormal Behaviour Detection in Video [article]

Olga Isupova, Danil Kuzin, Lyudmila Mihaylova
2016 arXiv   pre-print
Conventional posterior inference algorithms for this kind of models require processing of the whole data through several passes. It is computationally intractable for massive or sequential data.  ...  This paper proposes a novel dynamic Hierarchical Dirichlet Process topic model that considers the dependence between successive observations.  ...  Hierarchical Dirichlet Process Topic Model This kind of mixture models with a potentially infinite number of mixture components can be modelled with the Hierarchical Dirichlet Process (HDP) [18] .  ... 
arXiv:1606.08476v1 fatcat:oyulwt6zefc73osofeasnowc2u

Surface Estimation and Tracking using Sequential MCMC Methods for Video Based Rendering

Adam Bowen, Andrew Mullins, Roland Wilson, Nasir Rajpoot
2007 2007 IEEE International Conference on Image Processing  
mixture model analysis of the surface data.  ...  Video based rendering algorithms attempt to render videos of a scene from an arbitrary viewpoint, given a set of input video sequences taken from several fixed viewpoints.  ...  In this iterative process, the posterior distribution becomes the prior distribution for the next frame of video.  ... 
doi:10.1109/icip.2007.4379217 dblp:conf/icip/BowenMWR07 fatcat:cokcbdzyzjg57ngiqgnt7hxux4

Dual Sticky Hierarchical Dirichlet Process Hidden Markov Model and Its Application to Natural Language Description of Motions

Weiming Hu, Guodong Tian, Yongxin Kang, Chunfeng Yuan, Stephen Maybank
2018 IEEE Transactions on Pattern Analysis and Machine Intelligence  
In this paper, a new nonparametric Bayesian model called the dual sticky hierarchical Dirichlet process hidden Markov model (HDP-HMM) is proposed for mining activities from a collection of time series  ...  All the time series data are clustered. Each cluster of time series data, corresponding to a motion pattern, is modeled by an HMM.  ...  A distribution generated from a Dirichlet process corresponds to a restaurant.  ... 
doi:10.1109/tpami.2017.2756039 pmid:28952936 fatcat:6lyk2y3x3zforcjvqzlg7n4fba

Nonparametric Variational Auto-encoders for Hierarchical Representation Learning [article]

Prasoon Goyal, Zhiting Hu, Xiaodan Liang, Chenyu Wang, Eric Xing
2017 arXiv   pre-print
The resulting model induces a hierarchical structure of latent semantic concepts underlying the data corpus, and infers accurate representations of data instances.  ...  We apply our model in video representation learning.  ...  FA8702-15-D-0002 with Carnegie Mellon University for the operation of the Software Engineering Institute, a federally funded research and development center.  ... 
arXiv:1703.07027v2 fatcat:fllrzupobrfhjpjeejjgzgmve4

Bayesian Nonparametric Approaches to Abnormality Detection in Video Surveillance

Vu Nguyen, Dinh Phung, Duc-Son Pham, Svetha Venkatesh
2015 Annals of Data Science  
In particular, we employ the Infinite Hidden Markov Model and Bayesian Nonparametric Factor Analysis for stream data segmentation and pattern discovery.  ...  In data science, anomaly detection is the process of identifying the items, events or observations which do not conform to expected patterns in a dataset.  ...  One particular attractive approach is the Hierarchical Dirichlet Processes (HDP) [16] which posits the dependency among the group-level DPM by another Dirichlet process.  ... 
doi:10.1007/s40745-015-0030-3 fatcat:56wmpggdgraktipdvg5solnnp4

Nonparametric discovery of activity patterns from video collections

Michael C. Hughes, Erik B. Sudderth
2012 2012 IEEE Computer Society Conference on Computer Vision and Pattern Recognition Workshops  
We extend the earlier beta process HMM in two ways: adding data-driven MCMC moves to improve inference on realistic datasets, and using a hierarchical beta process HMM (HBP-HMM) to improve behavior sharing  ...  We propose a nonparametric framework based on the beta process for discovering temporal patterns within a heterogenous video collection.  ...  Acknowledgments The data used in this paper was obtained from kitchen.cs.cmu.edu and the data collection was funded in part by the National Science Foundation under Grant No. EEEC-0540865  ... 
doi:10.1109/cvprw.2012.6239170 dblp:conf/cvpr/HughesS12 fatcat:xg5bvpi3krelnbx7cgjztautv4

Using Bayesian Nonparametric Hidden Semi-Markov Models to Disentangle Affect Processes during Marital Interaction

William A. Griffin, Xun Li, Wael El-Deredy
2016 PLoS ONE  
Using extant data we developed a method of classifying and subsequently generating couple dynamics using a Hierarchical Dirichlet Process Hidden semi-Markov Model (HDP-HSMM).  ...  and affective micro-social processes.  ...  Analyzed the data: WAG XL. Contributed reagents/materials/analysis tools: WAG XL. Wrote the paper: WAG.  ... 
doi:10.1371/journal.pone.0155706 pmid:27187319 pmcid:PMC4871360 fatcat:2ycire6s55fypkhmrprs5zkkla

Temporal Unknown Incremental Clustering Model for Analysis of Traffic Surveillance Videos

Kelathodi Kumaran Santhosh, Debi Prosad Dogra, Partha Pratim Roy
2018 IEEE transactions on intelligent transportation systems (Print)  
Optimized scene representation is an important characteristic of a framework for detecting abnormalities on live videos.  ...  One of the challenges for detecting abnormalities in live videos is real-time detection of objects in a non-parametric way.  ...  The method works in the absence of complete training data. They have processed the data sequentially.  ... 
doi:10.1109/tits.2018.2834958 fatcat:jno7ekyzyndbjchnz3xd3j2l5i

Multi-camera open space human activity discovery for anomaly detection

Remi Emonet, Jagannadan Varadarajan, Jean-Marc Odobez
2011 2011 8th IEEE International Conference on Advanced Video and Signal Based Surveillance (AVSS)  
We address the discovery of typical activities in video stream contents and its exploitation for estimating the abnormality levels of these streams.  ...  Our contributions come from the following facets: i) the method is fully unsupervised and learns the activities from long term data; ii) the method is scalable and can efficiently handle the information  ...  We further process these features with a Hierarchical Dirichlet Process (HDP) [8] (in place of Probabilistic Latent Semantic Analysis, PLSA).  ... 
doi:10.1109/avss.2011.6027325 dblp:conf/avss/EmonetVO11 fatcat:4lrflhp57feozm63pxkyfeflte
« Previous Showing results 1 — 15 out of 1,330 results