16 Hits in 12.1 sec

A Topological Data Analysis Guided Fusion Algorithm: Mapper-Regularized Manifold Alignment

Jingliang Hu, Danfeng Hong, Yuanyuan Wang, Xiao Xiang Zhu
2019 IGARSS 2019 - 2019 IEEE International Geoscience and Remote Sensing Symposium  
To our best knowledge, this paper investigates for the first time semisupervised manifold alignment (SSMA) for the fusion of the hyperspectral image and PolSAR data.  ...  The paper also proposes a novel algorithm, MAPPER-regularized manifold alignment, which embeds the TDA into a semisupervised manifold alignment for the fusion of the hyperspectral image and PolSAR data  ...  The manifold alignment technique has successfully accomplished data fusion in both the machine learning field and remote sensing applications [8] [9] [10] .  ... 
doi:10.1109/igarss.2019.8898471 dblp:conf/igarss/HuH0Z19 fatcat:l6neki2bivappmg3uhl2t5nzta

Multimodal Remote Sensing Benchmark Datasets for Land Cover Classification with A Shared and Specific Feature Learning Model [article]

Danfeng Hong and Jingliang Hu and Jing Yao and Jocelyn Chanussot and Xiao Xiang Zhu
2021 arXiv   pre-print
To this end, we propose a shared and specific feature learning (S2FL) model.  ...  S2FL is capable of decomposing multimodal RS data into modality-shared and modality-specific components, enabling the information blending of multi-modalities more effectively, particularly for heterogeneous  ...  Acknowledgement The authors would like to the Hyperspectral Image Analysis group at the University of Houston and the IEEE GRSS DFC2013 for providing the University of Houston HS dataset.  ... 
arXiv:2105.10196v1 fatcat:hjm5sqvugfc5te4saylthvz4fm

Advances in Hyperspectral Image and Signal Processing: A Comprehensive Overview of the State of the Art

Pedram Ghamisi, Naoto Yokoya, Jun Li, Wenzhi Liao, Sicong Liu, Javier Plaza, Behnood Rasti, Antonio Plaza
2017 IEEE Geoscience and Remote Sensing Magazine  
Recent advances in airborne and spaceborne hyperspectral imaging technology have provided end users with rich spectral, spatial, and temporal information, which make a plethora of applications for the  ...  Hence, rigorous and innovative methodologies are required for hyperspectral image and signal processing and have become a center of attention for researchers worldwide.  ...  and Data Fusion Technical Committee for organizing the 2013 Data Fusion Contest.  ... 
doi:10.1109/mgrs.2017.2762087 fatcat:6ezzye7yyvacbouduqv2f2c7gi

2014 Index IEEE Transactions on Neural Networks and Learning Systems Vol. 25

2014 IEEE Transactions on Neural Networks and Learning Systems  
., +, TNNLS Apr. 2014 690-703 Learning Deep and Wide: A Spectral Method for Learning Deep Networks.  ...  ., +, TNNLS Mar. 2014 557-570 Error statistics Learning Deep and Wide: A Spectral Method for Learning Deep Networks.  ...  The Field of Values of a Matrix and Neural Networks. Georgiou, G.M., TNNLS Sep. 2014  ... 
doi:10.1109/tnnls.2015.2396731 fatcat:ztnfcozrejhhfdwg7t2f5xlype

Automatic assessment of average diaphragm motion trajectory from 4DCT images through machine learning

Guang Li, Jie Wei, Hailiang Huang, Carl Philipp Gaebler, Amy Yuan, Joseph O Deasy
2015 Biomedical engineering and physics express  
We have developed an effective motion extraction approach and a machine-learning-based algorithm to estimate the ADMT.  ...  The mean error in the predicted ADMT using leave-one-out method was 0.3 ± 1.9 mm for the left-side diaphragm and 0.0 ± 1.4 mm for the right-side diaphragm.  ...  The authors thank Dr G Mageras for sharing the amplitude-binning 4DCT images, and Drs O Dzyubak and R Kincaid for their assistance in initial 4DCT image processing.  ... 
doi:10.1088/2057-1976/1/4/045015 pmid:27110388 pmcid:PMC4840474 fatcat:5525dpcdjrfcneomytsnymeptq

A Review of Human Activity Recognition Methods

Michalis Vrigkas, Christophoros Nikou, Ioannis A. Kakadiaris
2015 Frontiers in Robotics and AI  
Many applications, including video surveillance systems, human-computer interaction, and robotics for human behavior characterization, require a multiple activity recognition system.  ...  Moreover, we provide a comprehensive analysis of the existing, publicly available human activity classification datasets and examine the requirements for an ideal human activity recognition dataset.  ...  ACKNOWLEDGMENTS This research was funded in part by the UH Hugh Roy and Lillie Cranz Cullen Endowment Fund.  ... 
doi:10.3389/frobt.2015.00028 fatcat:ywzq5ej2gbhatg62sp46t3usgi

2020 Index IEEE Transactions on Industrial Informatics Vol. 16

2020 IEEE Transactions on Industrial Informatics  
Gao, F., Data-Driven Two-Dimensional Deep Correlated Representation Learning for Nonlinear Batch Process Monitoring; TII April 2020 2839-2848 Jiang, S., see Li, Y., 1076-1085 Jiang, X., see Gong, K.,  ...  -5149 Meng, K., see Zhang, Y., TII July 2020 4390-4402 Meng, K., Jia, Y., Yang, H., Niu, F., Wang, Y., and Sun, D., Motion Planning and Robust Control for the Endovascular Navigation of a Microrobot  ...  ., +, TII Oct. 2020 6543-6552 Control charts Multivariate Nonparametric Control Chart With Storage Space.  ... 
doi:10.1109/tii.2021.3053362 fatcat:blfvdtsc3fdstnk6qoaazskd3i

Deep Learning on Attributed Graphs: A Journey from Graphs to Their Embeddings and Back [article]

Martin Simonovsky
2019 arXiv   pre-print
The goal of this thesis is to investigate architectures for DL on graphs and study how to transfer, adapt or generalize concepts that work well on sequential and image data to this domain.  ...  Third, we present GraphVAE, a graph generator allowing us to decode graphs with variable but upper-bounded number of nodes making use of approximate graph matching for aligning the predictions of an autoencoder  ...  Acknowledgements Foremost, I would like to express gratitude to my supervisor Nikos Komodakis for accepting me as his student and securing funding for the whole period, for useful discourses and insights  ... 
arXiv:1901.08296v1 fatcat:vp6uzgasnrgzpoaqv2dni5bega

Adversarial approaches to remote sensing image analysis [article]

Bejiga Mesay Belete
We apply this method for large-scale land cover classification and cross-sensor hyperspectral classification problems.  ...  The goal of domain adaptation is to learn a generic classifier for multiple related problems, thereby reducing the cost of labeling. To that end, we propose two methods.  ...  In [84] , two alignment methods were proposed for multi-temporal hyperspectral image classification. The first method uses source manifold as a prior to learn a joint manifold embedding.  ... 
doi:10.15168/11572_257100 fatcat:o3qb4wb4jjfbxnizkniqurc4pu

Learning With Multiple Views

Stefan Rüping, Tobias Scheffer
2005 Proceedings of the ICML 2005 Workshop on   unpublished
This paper is concerned with a real active learning experiment to bootstrap a named entity recognition system for a new domain of radio astronomical abstracts.  ...  We evaluate several committee-based metrics for quantifying the disagreement between classifiers built using multiple views, and demonstrate that the choice of metric can be optimised in simulation experiments  ...  of spectral clustering code at deepak/spectral/library.tgz.  ... 

Robust Video Content Analysis via Transductive Learning Methods

Ralph Ewerth, Mathematik Und Informatik, Bernd Freisleben
In addition, the variability of multimedia sources and content is enormous, and obviously this is also true for video databases.  ...  Hence, there is a need for the development of algorithms that work reliably and independently of the factors mentioned above -a robust video content 2005-2007 wären ohne den unermüdlichen Einsatz von Christian  ...  the high-level feature "charts" for different learning strategies.  ... 
doi:10.17192/z2008.0478 fatcat:mr7ilqdz6ndefp2plqfyeu5hqm

Machine learning for magnetic resonance image reconstruction and analysis

Chen Qin, Daniel Rueckert
Finally, deep learning approaches for image registration and its applications are investigated.  ...  Machine learning has great potentials to improve the entire medical imaging pipeline, providing support for clinical decision making and computer-aided diagnosis.  ...  In this chapter, we propose a novel unspervised registration method for aligning intrasubject multi-modal images, without the need of ground truth deformation fields, aligned multimodal image pairs or  ... 
doi:10.25560/79301 fatcat:bdyyeutojjas3cppkur6xn6aqe

On Fitting Finite Dirichlet Mixture Using ECM and MML [chapter]

Nizar Bouguila, Djemel Ziou
2005 Lecture Notes in Computer Science  
The system is analysed for different active periods corresponding to the service time of an M/G/1 queue and the busy period of an M/G/1 queue.  ...  Numerical results are illustrated for the models under consideration. 2 Pair-copula constructions of multiple dependence.  ...  First, the linear transformation is defined on the stiefel manifold (the manifold of orthnormal matrices).  ... 
doi:10.1007/11551188_19 fatcat:gtzi5u6emvfyxea3zbklobnvom

Special Issue: eGovernment in Balkan Countries

Matjaž Gams, Marjan Gusev, Božidar Kliček, Veljko Milutinovi´c, Milutinovi´c Brane, Vujiči´ Vujiči´c, Anton Železnikar, Matjaž Gams, Jožef Stefan, Drago Torkar, Jožef Stefan, Ivan Bratko (+38 others)
2007 unpublished
Basic prerequisites for further development of ICT have been established through state-level adoption of documents covering the Policy, Strategy and Action plan for the development of the information society  ...  Next, the e-Croatia 2007 Programme projects and areas for development are discussed, with the analysis of the achievements so far.  ...  Many thanks go to the students of FH JOANNEUM, Department of Information Design, IND04 and IND05 for their willingness in participation at the pre and post survey as well as for their work during the courses  ... 

Feature Interactions: The Next Generation

Sven Apel, Joanne Atlee, Luciano Baresi, Pamela Zave, Abraham Bernstein, Jan Leimeister, Natasha Noy, Cristina Sarasua, Elena Simperl, Dirk Kutscher, Taekyoung Kwon, Ignacio Solis (+8 others)
2014 unpublished
Büchler has set up the basic infrastructure for this and we will begin to populate it as we experiment with teaching EBD.  ...  The group thanks Bettina Berendt for her fruitful hints, comments, and discussion of the topics discussed by this working group. Acknowledgements.  ...  and align them with implications for the steps needed in every research process.  ... 
« Previous Showing results 1 — 15 out of 16 results