Filters








28,066 Hits in 3.1 sec

Nonlinear Prediction for Gaussian Mixture Image Models

J. Zhang, D. Ma
2004 IEEE Transactions on Image Processing  
In this paper, we derive an optimal predictor for an important class of non-Gaussian image models, the block-based multivariate Gaussian mixture model.  ...  When the images are modeled as Gaussian, the optimal predictor is linear and easy to obtain.  ...  Figueiredo for providing us with some of his codes for EM algorithm-based clustering.  ... 
doi:10.1109/tip.2004.828197 pmid:15648873 fatcat:t4kepxha2jgudiqw2qywm6f4ny

Improvement of the GMM‐AR classification of multiframe contrast ultrasound images using gaussian filter

Bilal Ghazal, Maha Khachab, Denis Friboulet, Chafic Mokbel, Christian Cachard
2008 Journal of the Acoustical Society of America  
Then, a Gaussian filter is applied to the prediction error distribution before classification by a Gaussian mixture model.  ...  We have applied a new approach based on the autoregressive model where an image of prediction errors is calculated in the first phase.  ...  We are graceful to the AUF (Agence Universitaire de la Francophonie) grant for supporting this work.  ... 
doi:10.1121/1.2933416 fatcat:6m5yutp4qngl3dhrdxtdkjb5s4

Detection of nonlinear mixtures using Gaussian processes: Application to hyperspectral imaging

T. Imbiriba, J. C. M. Bermudez, J.-Y. Tourneret, C. Richard
2014 2014 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP)  
The proposed technique is independent of nonlinear mixing mechanism, and therefore is not restricted to any prescribed nonlinear mixing model.  ...  This paper investigates the use of Gaussian processes to detect nonlinearly mixed pixels in hyperspectral images.  ...  Recently, Altmann et al [14] proposed a robust nonlinear mixture detector that does not use a parametric model for the nonlinear mixture.  ... 
doi:10.1109/icassp.2014.6855148 dblp:conf/icassp/ImbiribaBTR14 fatcat:xt3kxeblq5fehgrwxsrrr5c2de

Nonlinear hyperspectral unmixing using Gaussian processes

Y. Altmann, N. Dobigeon, J.-Y. Tourneret, S. McLaughlin
2013 2013 5th Workshop on Hyperspectral Image and Signal Processing: Evolution in Remote Sensing (WHISPERS)  
The first step of the proposed method estimates the abundance vectors for all the image pixels using a Gaussian process latent variable model.  ...  This paper presents an unsupervised algorithm for nonlinear unmixing of hyperspectral images.  ...  GAUSSIAN PROCESS REGRESSION This section studies a new endmember estimation strategy based on GP regression for nonlinear mixtures.  ... 
doi:10.1109/whispers.2013.8080729 dblp:conf/whispers/AltmannDTM13 fatcat:74kkdeht6bby5oufwn22oxti4i

From filters to features: Scale–space analysis of edge and blur coding in human vision

Mark A. Georgeson, Keith A. May, Tom C. A. Freeman, Gillian S. Hesse
2007 Journal of Vision  
The model predicts remarkably accurately our results on human perception of edge location and blur for a wide range of luminance profiles, including the surprising finding that blurred edges look sharper  ...  Here we use Gaussian scale-space theory to derive a multiscale model for edge analysis and we test it in perceptual experiments. At all scales there are two stages of spatial filtering.  ...  The pattern of deviation for the other models was fairly similar to the blur mixture experiment.  ... 
doi:10.1167/7.13.7 pmid:17997635 fatcat:ehgpdgribnauvov6vasd6pfox4

Real-time Body Tracking Using a Gaussian Process Latent Variable Model

Shaobo Hou, Aphrodite Galata, Fabrice Caillette, Neil Thacker, Paul Bromiley
2007 2007 IEEE 11th International Conference on Computer Vision  
In this paper, we present a tracking framework for capturing articulated human motions in real-time, without the need for attaching markers onto the subject's body.  ...  This is achieved by first obtaining a low dimensional representation of the training motion data, using a nonlinear dimensionality reduction technique called back-constrained GPLVM.  ...  Predictive Dynamic Model A predictive dynamic model is necessary for ensuring efficient propagation of particles and robust handling of ambiguous image evidence.  ... 
doi:10.1109/iccv.2007.4408946 dblp:conf/iccv/HouGCTB07 fatcat:wu2plmwurrenzi2fmiqa2dlarq

Hybrid linear and nonlinear complexity pursuit for blind source separation

Zhenwei Shi, Hongjuan Zhang, Zhiguo Jiang
2012 Journal of Computational and Applied Mathematics  
nonlinear predictability.  ...  In this paper, we propose a hybrid BSS method based on linear and nonlinear complexity pursuit, which combines three statistical properties of source signals: non-Gaussianity, linear predictability and  ...  Fig. 6 .Fig. 7 . 67 Simulation results for the mixture of two images and one Gaussian signal.  ... 
doi:10.1016/j.cam.2012.03.022 fatcat:w52axldrd5ekfmzrow362xvyii

Large Scale Variational Bayesian Inference for Structured Scale Mixture Models [article]

Young Jun Ko, Matthias Seeger
2012 arXiv   pre-print
We derive a large scale approximate Bayesian inference algorithm for linear models with non-factorial (latent tree-structured) scale mixture priors.  ...  Representing such prior knowledge in non-factorial latent tree models can boost performance of image denoising, inpainting, deconvolution or reconstruction substantially, beyond standard factorial "sparse  ...  Mixture potentials can have arbitrary super-Gaussian components. • A large scale double loop algorithm for Bayesian inference in these hybrid models.  ... 
arXiv:1206.6437v1 fatcat:o3qhatvwt5bj5cbil63yaesgru

Are Gaussian spectra a viable perceptual assumption in color appearance?

Yoko Mizokami, Michael A. Webster
2011 Optical Society of America. Journal A: Optics, Image Science, and Vision  
for a trichromatic visual system.  ...  However, such models do not accommodate nonlinearities in color appearance, such as the Abney effect.  ...  This behavior of the Gaussian model at longer wavelengths can also account for the breakdown of the Gaussian predictions for the Abney effect at longer wavelengths, as noted by Mizokami et al.  ... 
doi:10.1364/josaa.29.000a10 pmid:22330365 pmcid:PMC3281511 fatcat:22tympvtu5gjfacq62hyxczxqe

Origin and Function of Tuning Diversity in Macaque Visual Cortex

Robbe L.T. Goris, Eero P. Simoncelli, J. Anthony Movshon
2015 Neuron  
Shapley for helpful discussions and to Nicholas Priebe for sharing his data with us.  ...  ACKNOWLEDGMENTS This work was supported by NIH grants EY04440 and EY022428, the Howard Hughes Medical Institute, and postdoctoral fellowships from the Fund for Scientific Research of Flanders and the Belgian  ...  The full model (ordinate) predicts the measured tuning curves for the different mixture stimuli better than any of the restricted models (top row).  ... 
doi:10.1016/j.neuron.2015.10.009 pmid:26549331 pmcid:PMC4786576 fatcat:2ip7k6pfu5drxgdsjbipvnt5p4

A Sample-based Criterion for Unsupervised Learning of Complex Models beyond Maximum Likelihood and Density Estimation

Mani Manavalan, Praveen Kumar Donepudi
2016 ABC Journal of Advanced Research  
Finally, simulations of linear and nonlinear models on mixtures of Gaussians and ICA issues are used to evaluate our approach.  ...  Recoding models like ICA and projection pursuit, as well as generative models like Gaussian mixtures and Boltzmann machines, can be seen in this perspective.  ...  For 𝑓𝑤 𝑚 (𝑧 𝑖 ), we employed a 3-layer sigmoidal neural network for a nonlinear mixture model (z i ).  ... 
doi:10.18034/abcjar.v5i2.581 fatcat:24dwqtchxjg2zimnn2m6m4jlna

Nonlinear Prediction Based on Independent Component Analysis Mixture Modelling [chapter]

Gonzalo Safont, Addisson Salazar, Luis Vergara
2011 Lecture Notes in Computer Science  
This paper presents a new algorithm for nonlinear prediction based on independent component analysis mixture modelling (ICAMM).  ...  Nonlinear prediction based on independent component analysis mixture modelling. Abstract.  ...  The linear ICA method is extended in independent component analysis mixture modelling (ICAMM) to a kind of nonlinear ICA model, i.e., multiple ICA models are learned and weighted in a probabilistic manner  ... 
doi:10.1007/978-3-642-21498-1_64 fatcat:s5ocxvd4vfbzxgzxvhq5idaqma

Generalized Probabilistic U-Net for medical image segementation [article]

Ishaan Bhat, Josien P.W. Pluim, Hugo J. Kuijf
2022 arXiv   pre-print
For the LIDC-IDRI dataset, we show that using a mixture of Gaussians results in a statistically significant improvement in the generalized energy distance (GED) metric with respect to the standard Probabilistic  ...  We show that the choice of distribution affects the sample diversity of the predictions and their overlap with respect to the reference segmentations.  ...  We also observed that models using a Gaussian mixture distribution had better overlap but produced less diverse predictions than single Gaussian variants, for both axis-aligned and full covariance distributions  ... 
arXiv:2207.12872v1 fatcat:2ctyua57srhsnf6lkysxw645k4

Are Gaussian spectra a viable perceptual assumption in color appearance?

Y. Mizokami, M. Webster
2010 Journal of Vision  
for a trichromatic visual system.  ...  However, such models do not accommodate nonlinearities in color appearance, such as the Abney effect.  ...  This behavior of the Gaussian model at longer wavelengths can also account for the breakdown of the Gaussian predictions for the Abney effect at longer wavelengths, as noted by Mizokami et al.  ... 
doi:10.1167/10.7.399 fatcat:hzacgejoazdrfji44rkpa33z4y

3D-GMNet: Single-View 3D Shape Recovery as A Gaussian Mixture [article]

Kohei Yamashita, Shohei Nobuhara, Ko Nishino
2020 arXiv   pre-print
In this paper, we introduce 3D-GMNet, a deep neural network for 3D object shape reconstruction from a single image. As the name suggests, 3D-GMNet recovers 3D shape as a Gaussian mixture.  ...  We train 3D-GMNet end-to-end with single input images and corresponding 3D models by introducing two novel loss functions, a 3D Gaussian mixture loss and a 2D multi-view loss, which collectively enable  ...  We use 80% of the 3D models in ShapeNet for training, 10% for validation, and the rest for testing. Fig. 3 shows predicted 3D models.  ... 
arXiv:1912.04663v2 fatcat:6uwwkzup3zfwjjcsmq2jferpsy
« Previous Showing results 1 — 15 out of 28,066 results