Filters








633 Hits in 4.6 sec

Geometric Multimodal Deep Learning with Multi-Scaled Graph Wavelet Convolutional Network [article]

Maysam Behmanesh, Peyman Adibi, Mohammad Saeed Ehsani, Jocelyn Chanussot
2021 arXiv   pre-print
Several semi-supervised node classification experiments have been conducted on three popular unimodal explicit graph-based datasets and five multimodal implicit ones.  ...  The geometry-aware data analysis approaches provide these capabilities by implicitly representing data in various modalities based on their geometric underlying structures.  ...  In [18] , another multimodal manifold learning approach, called local signal expansion for joint diagonalization (LSEJD) was proposed, which uses the intrinsic local tangent spaces to expand the initial  ... 
arXiv:2111.13361v1 fatcat:6dluczwatfdcrn6lkvfwauty3m

Cross-Modal and Multimodal Data Analysis Based on Functional Mapping of Spectral Descriptors and Manifold Regularization [article]

Maysam Behmanesh, Peyman Adibi, Jocelyn Chanussot, Sayyed Mohammad Saeed Ehsani
2021 arXiv   pre-print
The second method is a manifold regularized multimodal classification based on pointwise correspondences (M^2CPC) used for the problem of multiclass classification of multimodal heterogeneous, which the  ...  For this purpose, first, we identify the localities of each manifold by extracting local descriptors via spectral graph wavelet signatures (SGWS).  ...  local tangent spaces for signal expansion of multimodal problems.  ... 
arXiv:2105.05631v1 fatcat:64uqblnzb5btnnhkeuzortajtq

ModDrop: adaptive multi-modal gesture recognition [article]

Natalia Neverova and Christian Wolf and Graham W. Taylor and Florian Nebout
2015 arXiv   pre-print
We present a method for gesture detection and localisation based on multi-scale and multi-modal deep learning.  ...  Futhermore, the proposed ModDrop training technique ensures robustness of the classifier to missing signals in one or several channels to produce meaningful predictions from any number of available modalities  ...  The major contributions of the present work are the following: We (i) develop a deep learning-based multimodal and multi-scale framework for gesture detection, localization and recognition, which can be  ... 
arXiv:1501.00102v2 fatcat:p5a7xrp6zfc4jf2tuihvoq2j74

ModDrop: Adaptive Multi-Modal Gesture Recognition

Natalia Neverova, Christian Wolf, Graham Taylor, Florian Nebout
2016 IEEE Transactions on Pattern Analysis and Machine Intelligence  
We present a method for gesture detection and localisation based on multi-scale and multi-modal deep learning.  ...  Futhermore, the proposed ModDrop training technique ensures robustness of the classifier to missing signals in one or several channels to produce meaningful predictions from any number of available modalities  ...  The major contributions of the present work are the following: We (i) develop a deep learning-based multimodal and multi-scale framework for gesture detection, localization and recognition, which can be  ... 
doi:10.1109/tpami.2015.2461544 fatcat:d6dp5cdmfbgd3awuwfimtqrqdm

Table of Contents

2021 IEEE Transactions on Signal Processing  
Zhao Geometric Multimodal Learning Based on Local Signal Expansion for Joint Diagonalization . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .  ...  Dokmanić Efficient Sensor Placement for Signal Reconstruction Based on Recursive Methods . . . . . . . ..B. Li, H. Liu, and R.  ... 
doi:10.1109/tsp.2021.3136800 fatcat:zhf46mb3rbdlnnh3u2xizgxof4

Table of Contents

2021 IEEE Transactions on Signal Processing  
Sayed Geometric Multimodal Learning Based on Local Signal Expansion for Joint Diagonalization . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .  ...  Erez Efficient Sensor Placement for Signal Reconstruction Based on Recursive Methods . . . . . . . . . B. Li, H. Liu, and R.  ... 
doi:10.1109/tsp.2021.3136798 fatcat:kzkdhzcz3fgx3jv6gfjofooseq

Recovering Hidden Components in Multimodal Data with Composite Diffusion Operators [article]

Tal Shnitzer, Mirela Ben-Chen, Leonidas Guibas, Ronen Talmon, Hau-Tieng Wu
2018 arXiv   pre-print
Based on these new operators, efficient low-dimensional representations can be constructed for such data, which characterize the common structures and the differences between the manifolds underlying the  ...  In this paper, we address the problem using manifold learning, where the data from each modality is assumed to lie on some manifold.  ...  For example, methods for spectral clustering of multimodal data based on kernel manipulation are presented in [40, 20, 10] .  ... 
arXiv:1808.07312v1 fatcat:fvpiildi4zalveym2pkfmgbyfa

Harmonic Alignment [article]

Jay S. Stanley III, Scott Gigante, Guy Wolf, Smita Krishnaswamy
2020 arXiv   pre-print
We propose a novel framework for combining datasets via alignment of their intrinsic geometry.  ...  This alignment is obtained by relating the expansion of data features in harmonics derived from diffusion operators defined over each dataset.  ...  One of the earliest attempts at manifold alignment (in particular, with no point correspondence), was presented in [11] , which proposes a linear method based on embedding a joint graph built over both  ... 
arXiv:1810.00386v4 fatcat:uuqznvio25h4bowinyy2h7hi6y

Probabilistic visual learning for object representation

B. Moghaddam, A. Pentland
1997 IEEE Transactions on Pattern Analysis and Machine Intelligence  
We present an unsupervised technique for visual learning, which is based on density estimation in high-dimensional spaces using an eigenspace decomposition.  ...  Two types of density estimates are derived for modeling the training data: a multivariate Gaussian (for unimodal distributions) and a Mixture-of-Gaussians model (for multimodal distributions).  ...  We would also like to thank Wasiuddin Wahid for his hard work in helping us participate in the FERET testing.  ... 
doi:10.1109/34.598227 fatcat:is6hvdhogrgixhcpymhs7azb4e

A graph representation based on fluid diffusion model for multimodal data analysis: theoretical aspects and enhanced community detection [article]

Andrea Marinoni and Christian Jutten and Mark Girolami
2021 arXiv   pre-print
Nevertheless, classic graph signal processing is based on a model for information propagation that is configured according to heat diffusion mechanism.  ...  In this paper, we introduce a novel model for graph definition based on fluid diffusion.  ...  Moreover, for comparison, we used a method based on ensemble learning approach [64] .  ... 
arXiv:2112.04388v1 fatcat:2dqk4wws5vc7plbcwhlr4alhya

The shape of fuzzy sets in adaptive function approximation

S. Mitaim, B. Kosko
2001 IEEE transactions on fuzzy systems  
We present a method for constructing such unfactorable joint sets from scalar distance measures.  ...  No one set shape emerges as the best shape. The sinc function often does well and has a tractable learning law. But its undulating sidelobes may have no linguistic meaning.  ...  10 epochs of learning, (c) after 500 epochs, and (d) the sinc set converges after 3000 epochs. of all those local minima and maxima in such a multimodal set function.  ... 
doi:10.1109/91.940974 fatcat:bzfwllp4drfnzbgoiy455lojgy

Learning Deep and Wide: A Spectral Method for Learning Deep Networks

Ling Shao, Di Wu, Xuelong Li
2014 IEEE Transactions on Neural Networks and Learning Systems  
temporally local features whilst maintaining the holistic geometric information.  ...  The experimental results on bi-modal time series data, i.e., audio and skeletal joints data, show that the multimodal DBN+HMM framework can learn a good model of the joint space of multiple sensory inputs  ...  Result: -frame based local label where y i ∈ {C * S + 1} with C is the number of class, S is the number of hidden states for each class, 1 as ergodic state.  ... 
doi:10.1109/tnnls.2014.2308519 pmid:25420251 fatcat:4mnl6tv2xnf3jpzwhp76cvl4ti

Multimodal Classification of Remote Sensing Images: A Review and Future Directions

Luis Gomez-Chova, Devis Tuia, Gabriele Moser, Gustau Camps-Valls
2015 Proceedings of the IEEE  
We also highlight the most recent advances, which exploit synergies with machine learning and signal processing: sparse methods, kernel-based fusion, 2 Markov modeling, and manifold alignment.  ...  In this paper, we provide a taxonomical view of the field and review the current methodologies for multimodal classification of remote sensing images.  ...  ACKNOWLEDGEMENTS The authors would like to thank DigitalGlobe Inc. for the optical data on Rio and Haiti, and the Italian Space Agency for the SAR data on Haiti.  ... 
doi:10.1109/jproc.2015.2449668 fatcat:gaficd2bcrbshcrds3a2wfa25a

MultiBench: Multiscale Benchmarks for Multimodal Representation Learning [article]

Paul Pu Liang, Yiwei Lyu, Xiang Fan, Zetian Wu, Yun Cheng, Jason Wu, Leslie Chen, Peter Wu, Michelle A. Lee, Yuke Zhu, Ruslan Salakhutdinov, Louis-Philippe Morency
2021 arXiv   pre-print
Learning multimodal representations involves integrating information from multiple heterogeneous sources of data.  ...  MultiBench introduces impactful challenges for future research, including scalability to large-scale multimodal datasets and robustness to realistic imperfections.  ...  We believe that multimodal models are able to successfully rely on the other modality when one is  ... 
arXiv:2107.07502v2 fatcat:ls47dr7lpfhkbfry4r6dtqjtua

Sparsity-Driven Synthetic Aperture Radar Imaging: Reconstruction, autofocusing, moving targets, and compressed sensing

Mujdat Cetin, Ivana Stojanovic, Ozben Onhon, Kush Varshney, Sadegh Samadi, William Clem Karl, Alan S. Willsky
2014 IEEE Signal Processing Magazine  
He is on the editorial board of Digital Signal Processing and a member of the IEEE Signal Processing Society Machine Learning for Signal Processing Technical Committee.  ...  (d) A composite joint subaperture image imposing piecewise smoothness in angular scattering. (e) A composite joint subaperture image based on an overcomplete dictionary for angular scattering.  ...  The cumulative coherence provides an upper bound on the % t -mutual coherence. Note that ( ) H t% n  ... 
doi:10.1109/msp.2014.2312834 fatcat:47honwuywjgvnazjflqllmu7fm
« Previous Showing results 1 — 15 out of 633 results