657 Hits in 9.2 sec

Gaussian-weighted Jensen–Shannon divergence as a robust fitness function for multi-model fitting

Kai Zhou, Karthik Mahesh Varadarajan, Michael Zillich, Markus Vincze
2013 Machine Vision and Applications  
In this paper, we present a novel model evaluation function based on Gaussian-weighted Jensen-Shannon divergence, and integrate into a particle swarm optimization (PSO) framework using ring topology.  ...  Model fitting is a fundamental component in computer vision for salient data selection, feature extraction and data parameterization.  ...  Gaussian-weighted Jensen-Shannon divergence This section describes the details of the proposed evaluation function using Gaussian-weighted Jensen-Shannon divergence [18] .  ... 
doi:10.1007/s00138-013-0513-1 fatcat:splhbaeyy5eypioy5qsxc2idmy

Robust Face Alignment by Multi-order High-precision Hourglass Network [article]

Jun Wan, Zhihui Lai, Jun Liu, Jie Zhou, Can Gao
2020 arXiv   pre-print
To address the alignment problem for faces with extremely large poses and heavy occlusions, this paper proposes a heatmap subpixel regression (HSR) method and a multi-order cross geometry-aware (MCG) model  ...  At the same time, the MCG model is able to use the proposed multi-order cross information to learn more discriminative representations for enhancing facial geometric constraints and context information  ...  Acknowledgements We thank the reviewers for the suggestions and the NSCC for computing resources.  ... 
arXiv:2010.08722v1 fatcat:53utkvxrrffe5md7lv2pirquaq

Stratified regularity measures with Jensen-Shannon divergence

Kazunori Okada, Senthil Periaswamy, Jinbo Bi
2008 2008 IEEE Computer Society Conference on Computer Vision and Pattern Recognition Workshops  
Jensen-Shannon divergence is used to compute a set-similarity of intensity distributions derived from stratified data.  ...  We prove that derived regularity measures form a continuum as a function of the stratification's granularity and also upper-bounded by the Shannon entropy.  ...  Jensen-Shannon divergence (JS) was proposed by Lin [16] as a new distributional similarity function.  ... 
doi:10.1109/cvprw.2008.4563020 dblp:conf/cvpr/OkadaPB08 fatcat:ychcywqkz5g3hk7n3quvq6msyi

An Optimal Segmentation Method Using Jensen–Shannon Divergence via a Multi-Size Sliding Window Technique

Qutaibeh Katatbeh, José Martínez-Aroza, Juan Gómez-Lopera, David Blanco-Navarro
2015 Entropy  
The presented method computes the Jensen-Shannon divergence of the normalized grayscale histogram of a set of multi-sized double sliding windows over the entire image.  ...  In this paper we develop a new procedure for entropic image edge detection.  ...  Jensen-Shannon Divergence as an Edge Detector In this section, we will explain how to use unweighted Jensen-Shannon divergence to detect edges.  ... 
doi:10.3390/e17127858 fatcat:bzdiz7lmkfabjpyd5gmxhykyue

A Robust Algorithm for Characterizing Anisotropic Local Structures [chapter]

Kazunori Okada, Dorin Comaniciu, Navneet Dalal, Arun Krishnan
2004 Lecture Notes in Computer Science  
This paper proposes a robust estimation and validation framework for characterizing local structures in a positive multi-variate continuous function approximated by a Gaussian-based model.  ...  To this goal, it unifies robust statistical estimation for parametric model fitting and multi-scale analysis based on continuous scale-space theory.  ...  Acknowledgments The authors wish to thank Visvanathan Ramesh from Siemens Corporate Research for stimulating discussions, Alok Gupta from CAD group, Siemens Medical Solutions, for his support and encouragement  ... 
doi:10.1007/978-3-540-24670-1_42 fatcat:e6v3agnikvdgxer25oqkxgi5pa

Robust anisotropic Gaussian fitting for volumetric characterization of Pulmonary nodules in multislice CT

K. Okada, D. Comaniciu, A. Krishnan
2005 IEEE Transactions on Medical Imaging  
We propose a novel multi-scale joint segmentation and model fitting solution which extends the robust mean shift-based analysis to the linear scale-space theory.  ...  by robustly and efficiently fitting an anisotropic Gaussian intensity model.  ...  , and Jonathan Stoeckel from CAD group, Siemens Medical Solutions, for providing us the volume visualization tool and for his valuable technical supports.  ... 
doi:10.1109/tmi.2004.843172 pmid:15754991 fatcat:vof53t52crcbfnpyitj23l4hsq

Applications of entropic spanning graphs

A.O. Hero, Bing Ma, O.J.J. Michel, J. Gorman
2002 IEEE Signal Processing Magazine  
The «-entropy converges to the Shannon entropy Ê ´Þµ Ð Ò ´Þµ Þ as « ½. A related quantity is the «-divergence between two feature densities ½ and ¼ of order « ¾´¼ ½µ [5], [6], [7]  ...  Here the relevant notion of entropy is the «-entropy of the feature probability density , also known as Rényi entropy, which for probability densities is defined as [5] À «´ µ ½ ½ « ÐÒ «´Þ µ Þ for « ¾´  ...  As an illustrative example consider the case where ¼ and are multivariate Gaussian densities. The KL divergence for such a Gaussian feature model was adopted in [32] , [3] .  ... 
doi:10.1109/msp.2002.1028355 fatcat:plragdoykndwbgzfluthun3k2m

Anisotropic Scale Selection, Robust Gaussian Fitting, and Pulmonary Nodule Segmentation in Chest CT Scans [chapter]

Kazunori Okada
2011 Multi Modality State-of-the-Art Medical Image Segmentation and Registration Methodologies  
The theory combines two distinct concepts for generic data analysis: automatic scale selection and robust Gaussian model fitting.  ...  This chapter demonstrates how the resulting novel concept of anisotropic scale selection gives a useful and robust solution to the Gaussian fitting problem used as a part of our robust nodule segmentation  ...  Jensen-Shannon divergence is a natural extension of the pair-wise Kullback-Leibler divergence to describe similarity among a set of distributions [83] .  ... 
doi:10.1007/978-1-4419-8195-0_3 fatcat:sva6y55wvnbmlaassfpnnfu24i

Shape from semantic segmentation via the geometric Rényi divergence

Tatsuro Koizumi, William A. P. Smith
2021 2021 IEEE Winter Conference on Applications of Computer Vision (WACV)  
We propose a novel loss function based on a probabilistic, vertex-wise projection of the 3D model to the image plane.  ...  We represent both these projections and pixel labels as mixtures of Gaussians and compute the discrepancy between the two based on the geometric Rényi divergence.  ...  Smith was supported by a Royal Academy of Engineering/The Leverhulme Trust Senior Research Fellowship.  ... 
doi:10.1109/wacv48630.2021.00236 fatcat:zfgyjh4jx5bxhl3fpxhbiaat4i

Some Statistic and Information-theoretic Results on Arithmetic Average Density Fusion [article]

Tiancheng Li, Yue Xin, Yan Song, Enbin Song, Hongqi Fan
2021 arXiv   pre-print
The best fit of the mixture is formulated as a max-min problem, proving the sub-optimality of the AA fusion. Linear Gaussian models are considered for illustration and simulation comparison.  ...  Finite mixture such as the Gaussian mixture is a flexible and powerful probabilistic modeling tool for representing the multimodal distribution widely involved in many estimation and learning problems.  ...  Then, we discuss principles for fusing weight design. Some of the results are not limited to linear Gaussian models, but we use linear Gaussian models for illustration. We briefly conclude in Sec.  ... 
arXiv:2110.01440v3 fatcat:ni2i3dcxwrg4nnf4ok5rc4jvpm

A Deep Drift-Diffusion Model for Image Aesthetic Score Distribution Prediction [article]

Xin Jin, Xiqiao Li, Heng Huang, Xiaodong Li, Xinghui Zhou
2020 arXiv   pre-print
In recent years, the target representation of image aesthetic quality has changed from a one-dimensional binary classification label or numerical score to a multi-dimensional score distribution.  ...  In this paper, we propose a Deep Drift-Diffusion (DDD) model inspired by psychologists to predict aesthetic score distribution from images.  ...  Stochastic gradient descent is used to train our model with a mini-batch size of 48 images, a momentum of 0.9, a gamma of 0.5 and a weight decay of 0.0005. The max number of iterations is 120000.  ... 
arXiv:2010.07661v1 fatcat:3f72ujdyaff5fm6msvn3dpjnxa

Comparison of Redundancy and Relevance Measures for Feature Selection in Tissue Classification of CT Images [chapter]

Benjamin Auffarth, Maite López, Jesús Cerquides
2010 Lecture Notes in Computer Science  
Applied measures of correlation or distributional similarity for redunancy and relevance include Kolmogorov-Smirnov (KS) test, Spearman correlations, Jensen-Shannon divergence, and the sign-test.  ...  VDM was very good in our experiments as both redundancy and relevance measure. Jensen-Shannon and the sign-test are good redundancy measure alternatives and FC is a good relevance measure alternative.  ...  As for the redundancy measures, the Jensen-Shannon Divergence, RVDM, and the sign-test were good. RFC which is based on the relevance measure FC may have been too simple.  ... 
doi:10.1007/978-3-642-14400-4_20 fatcat:qgvij3a2gfejlhtkd53bz2crbq

GC Composition of the Human Genome: In Search of Isochores

Netta Cohen, Tal Dagan, Lewi Stone, Dan Graur
2005 Molecular biology and evolution  
We found that a four-family model of putative isochores is the most parsimonious multi-Gaussian model that can be fitted to the empirical data.  ...  The isochore theory, proposed nearly three decades ago, depicts the mammalian genome as a mosaic of long, fairly homogeneous genomic regions that are characterized by their guanine and cytosine (GC) content  ...  We thank Samuel Braunstein and Giddy Landan for numerous discussions and for their critical reading of the manuscript.  ... 
doi:10.1093/molbev/msi115 pmid:15728737 fatcat:ceztxsrtmzbjlhfuqthhlnb3ze

Probabilistic change detection and visualization methods for the assessment of temporal stability in biomedical data quality

Carlos Sáez, Pedro Pereira Rodrigues, João Gama, Montserrat Robles, Juan M. García-Gómez
2014 Data mining and knowledge discovery  
This work establishes the temporal stability as a data quality dimension and proposes new methods for its assessment based on a probabilistic framework.  ...  First, a probabilistic change detection algorithm is proposed based on the Statistical Process Control of the posterior Beta distribution of the Jensen-Shannon distance, with a memoryless forgetting mechanism  ...  Gregor Stiglic, from the Univeristy of Maribor, Slovenia, for his support on the NHDS data.  ... 
doi:10.1007/s10618-014-0378-6 fatcat:bmj7luikffachoj7op7vvwrqyq

Precise Segmentation of COVID-19 Infected Lung from CT Images Based on Adaptive First-Order Appearance Model with Morphological/Anatomical Constraints

Ahmed Sharafeldeen, Mohamed Elsharkawy, Norah Saleh Alghamdi, Ahmed Soliman Soliman, Ayman El-Baz
2021 Sensors  
To accurately model the distribution of Hounsfield scale values within both chest and lung regions, a new probabilistic model is developed that depends on a linear combination of Gaussian (LCG).  ...  Moreover, we modified the conventional expectation-maximization (EM) algorithm to be run in a sequential way to estimate both the dominant Gaussian components (one for the lung region and one for the chest  ...  Note that JSD stands for Jensen-Shannon divergence.  ... 
doi:10.3390/s21165482 pmid:34450923 pmcid:PMC8399192 fatcat:mnsebvyzazab3gdraknq3kj2wa
« Previous Showing results 1 — 15 out of 657 results