30,832 Hits in 4.4 sec

Using latent-structure to detect objects on the web

Luciano Barbosa, Juliana Freire
2010 Procceedings of the 13th International Workshop on the Web and Databases - WebDB '10  
In this paper, we address the problem of identifying pages that contain objects with a latent structure, i.e., the structure is implicitly represented in the page.  ...  These rules can then be used to detect the presence of these objects in new, unseen pages. Our approach has several advantages when compared against learning-based text classifiers.  ...  Our research has been funded by the National Science Foundation grants IIS-0905385, IIS-0844546, IIS-0746500, CNS-0751152, IIS-0713637, the Department of Energy, and an IBM Faculty Award.  ... 
doi:10.1145/1859127.1859138 dblp:conf/webdb/BarbosaF10 fatcat:tgdgue7akffnnbn7aq663iu3gu

Mining Deep And-Or Object Structures via Cost-Sensitive Question-Answer-Based Active Annotations [article]

Quanshi Zhang, Ying Nian Wu, Hao Zhang, Song-Chun Zhu
2019 arXiv   pre-print
The AOG explicitly represents object categories, poses/viewpoints, parts, and detailed structures within the parts in a compositional hierarchy.  ...  ., labeling parts on 3--10 training objects for each category) and achieved better performance than baseline methods.  ...  To reduce the effects of object detection on the APP, we detect the object within the image region of [c w ± w] and [c h ± h], where w/h/(c w , c h ) indicates the width/height/center of the true object  ... 
arXiv:1708.03911v3 fatcat:enuowzz2ojbcdfy6gqhffhnmqi

Leveraging relational autocorrelation with latent group models

Jennifer Neville, David Jensen
2005 Proceedings of the 4th international workshop on Multi-relational mining - MRDM '05  
We evaluate performance on three relational classification tasks and show that LGM outperforms models that ignore latent group structure when there is little known information with which to seed inference  ...  Recent research has explored the use of collective inference techniques to exploit this phenomenon.  ...  In these domains, LGMs could be used to detect group formation and use a few hand-labeled examples to seed inference about the classifications of new group members.  ... 
doi:10.1145/1090193.1090201 fatcat:6tlnr2saibdfvgxlapfik652ay

Web-Scale Multimedia Information Networks

Guo-Jun Qi, Min-Hsuan Tsai, Shen-Fu Tsai, Liangliang Cao, Thomas S. Huang
2012 Proceedings of the IEEE  
| The abundance of multimedia data on the Web presents both challenges (how to annotate, search, and mine) and opportunities (crawling the Web to create large structured multimedia data bases which can  ...  be used to do inference effectively).  ...  For example, Cheung and Zakhor [17] used metadata extracted from textual and hyperlink information to detect similar videos on the Web. Magalhaes et al.  ... 
doi:10.1109/jproc.2012.2201909 fatcat:4hcia4agvbaija2kx5ynj2esse

Weakly-Supervised Video Summarization Using Variational Encoder-Decoder and Web Prior [chapter]

Sijia Cai, Wangmeng Zuo, Larry S. Davis, Lei Zhang
2018 Lecture Notes in Computer Science  
To leverage the plentiful webcrawled videos to improve the performance of video summarization, we present a generative modelling framework to learn the latent semantic video representations to bridge the  ...  Experiments conducted on the challenging datasets CoSum and TVSum demonstrate the superior performance of the proposed VESD to existing state-of-the-art methods.  ...  We first use the features of segmented shots on web videos to pre-train a VAE module whose dimension of the latent variable is set to 256.  ... 
doi:10.1007/978-3-030-01264-9_12 fatcat:oa6tku77hvgotgmbibxzcmnhhq

Object Detection Using Strongly-Supervised Deformable Part Models [chapter]

Hossein Azizpour, Ivan Laptev
2012 Lecture Notes in Computer Science  
Deformable part-based models [1, 2] achieve state-of-the-art performance for object detection, but rely on heuristic initialization during training due to the optimization of non-convex cost function.  ...  We demonstrate significant improvements in detection performance compared to the LSVM [1] and the Poselet [3] object detectors.  ...  Acknowledgement.This work was supported by the VINST project funded by Swedish Foundation for Strategic Research (SSF), by the EIT ICT labs, by the Quaero program funded by OSEO and ERC grant VideoWorld  ... 
doi:10.1007/978-3-642-33718-5_60 fatcat:v2dnpq3d7bgyfdng6qfzuwwxge

Editorial: Special Issue on Probabilistic Models for Image Understanding

Bill Triggs, Christopher K. I. Williams
2010 International Journal of Computer Vision  
Li Li-Jia and Fei-Fei Li describe an online Bayesian learning method for image classification using large image collections on the web, which was tested on Caltech-101 and other datasets.  ...  , e.g. by exploiting latent structure or reusing components or priors.  ...  Williams, John Winn, and Andrew Zisserman give a detailed report on the results of the 2007 PASCAL Visual Object Classes Challenge with respect to the classification and detection problems.  ... 
doi:10.1007/s11263-010-0328-8 fatcat:ev5vvqr64ba6hpguq5nx6tke3q

Latent: a flexible data collection tool to research human behaviour in the context of web navigation

Catia Cepeda, Ricardo Tonet, Daniel Noronha Osorio, Hugo P. Silva, Edouard Battegay, Marcus Cheetham, Hugo Gamboa
2019 IEEE Access  
In this paper, we propose a novel tool in the form of a web browser extension, referred to as Latent, which can be used to simultaneously capture information from different sources while users interact  ...  While researchers and usability testing services do use these sources to analyze human behavior and user experience, access to the diverse range of other potentially useful data available during web-based  ...  (Catia Cepeda and Ricardo Tonet contributed equally to this work.)  ... 
doi:10.1109/access.2019.2916996 fatcat:r2wy2s737zbwpacrit6oe3tptu

Clustering based on Random Graph Model embedding Vertex Features [article]

Hugo Zanghi, Stevenn Volant, Christophe Ambroise
2009 arXiv   pre-print
In this paper, we provide a clustering algorithm exploiting both types of data based on a statistical model with latent structure characterizing each vertex both by a vector of features as well as by its  ...  Most techniques for clustering graph vertices just use the topology of connections ignoring informations in the vertices features.  ...  We use a classical Potts model on the latent structure which encourage spatial smoothing of the cluster.  ... 
arXiv:0910.2107v1 fatcat:7a22auh7xzbpdf3johvcbalffu


Pierre Andrews, Vanessa Murdock, Adam Rae, Francesco De Natale, Sven Buschbeck, Anthony Jameson, Kerstin Bischoff, Claudiu S. Firan, Claudia Niederée, Vasileios Mezaris, Spiros Nikolopoulos
2012 Proceedings of the 21st international conference companion on World Wide Web - WWW '12 Companion  
The idea of the European project GLOCAL is to use events as the central concept for search, organization and combination of multimedia content from various sources.  ...  goals, so as to be able to compare the associated media.  ...  In order to mitigate these problems, we automatically detect and learn about the points of interest (POIs) from text on the web by using a conditional random fields approach to sequence tagging that allows  ... 
doi:10.1145/2187980.2188013 dblp:conf/www/AndrewsNBJBFNMNMR12 fatcat:zhksq5dnorexvmf5ynpd4ez4iq

Estimation of sparse directed acyclic graphs for multivariate counts data

Sung Won Han, Hua Zhong
2016 Biometrics  
In searching for the solution, we use iterative optimization procedures to estimate the adjacency matrix and the variance matrix of the latent variables.  ...  Under the assumption that the count data follow the Poisson log-normal distribution, this paper provides an L 1 -penalized likelihood framework and an efficient search algorithm to estimate the structure  ...  In the next section, we use Linear Structural Equations to model the DAG structures on latent variable X.  ... 
doi:10.1111/biom.12467 pmid:26849781 pmcid:PMC4975686 fatcat:umyzmgfxnbas5b2i33wrrojgom

Empirical Evaluation of Web Based Alias Extraction Techniques

Selvaperumal, Suruliandi, Dhiliphan Rajkumar
2015 International Journal of Computer and Electrical Engineering  
Alias names are useful in a wide range of processes like information retrieval, opinion mining and terrorist activities detection and so on.  ...  Alias name extraction is the process of extracting surnames for the known name in the web.  ...  "Name nickname *" 0.234 This method mainly relies on anchor text graph mined from the web. For implementing the above work, instead of anchor text graph, web structure graph was constructed.  ... 
doi:10.17706/ijcee.2015.7.4.261-274 fatcat:fkzrgqzeynd5xhntgq3p2bpadu

Tuberculosis Control in South African Gold Mines: Mathematical Modeling of a Trial of Community-Wide Isoniazid Preventive Therapy

Emilia Vynnycky, Tom Sumner, Katherine L. Fielding, James J. Lewis, Andrew P. Cox, Richard J. Hayes, Elizabeth L. Corbett, Gavin J. Churchyard, Alison D. Grant, Richard G. White
2015 American Journal of Epidemiology  
lack of detectable population-level impact. 2) The optimized implementation increased impact by only 10%. 3) Implementing additional interventions individually and in combination led to up to 30% and  ...  We found the following: 1) The model suggests that a small proportion of latent infections among human immunodeficiency virus-positive people were cured, which could have been a key factor explaining the  ...  We thank the thousands of mine employees who consented to take part in this study.  ... 
doi:10.1093/aje/kwu320 pmid:25792607 pmcid:PMC4388015 fatcat:xgu3tvjq4zhqdnfq6ukrzpdp4q

Canonical contextual distance for large-scale image annotation and retrieval

Hideki Nakayama, Tatsuya Harada, Yasuo Kuniyoshi
2009 Proceedings of the First ACM workshop on Large-scale multimedia retrieval and mining - LS-MMRM '09  
In the experiment, we show that our method outperforms previous works using the standard Corel benchmark. Next, we verify our method by applying it to 3.5 million web images.  ...  Using the probabilistic structure, our similarity measure can reflect both appearance and semantics of samples.  ...  Using this structure, we derive the posterior probability of a sample in the latent space.  ... 
doi:10.1145/1631058.1631062 dblp:conf/mm/NakayamaHK09 fatcat:c3fvflp5hzasjpsfcdbxdl6h4u

Semantic Divergence Based Evaluation of Web Service Communities

Hafida Naim, Mustapha Aznag, Mohamed Quafafou, Nicolas Durand
2016 2016 IEEE International Conference on Services Computing (SCC)  
The number of community detection algorithms is growing continuously adopting a topological based approach to discover optimal subgraphs or communities.  ...  In this paper, we propose a new method combining both topology and semantic to evaluate and rank community detection algorithms.  ...  Our goal is not to use the semantic of web services and to combine it with the structure of the graph to find homogeneous web services communities like [12] .  ... 
doi:10.1109/scc.2016.101 dblp:conf/IEEEscc/NaimAQD16 fatcat:npufrbqrhzgu7ictbipuottunq
« Previous Showing results 1 — 15 out of 30,832 results