11,818 Hits in 6.8 sec

Machine Learning Paradigm towards Content Based Image Retrieval on High Resolution Satellite Images

In this paper, SVM based on random subspace and boosting ensemble learning is proposed for very high resolution satellite image retrieval.  ...  In order to retrieve high resolution satellite images Support Vector Machine (SVM) a machine learning paradigm is helpful for learning process and for pattern recognition and classification; ensemble methods  ...  Partitioning the query image into data points using random subspace develop the performance results. Every sub-classifier is learned comparable image subspace.  ... 
doi:10.35940/ijitee.b1104.1292s219 fatcat:phvtsg6kmnd45ebf6fisgwev6i

Human Age Estimation With Regression on Discriminative Aging Manifold

Yun Fu, Thomas S. Huang
2008 IEEE transactions on multimedia  
Through the manifold method of analysis on face images, the dimensionality redundancy of the original image space can be significantly reduced with subspace learning.  ...  In this paper, we demonstrate that such aging patterns can be effectively extracted from a discriminant subspace learning algorithm and visualized as distinct manifold structures.  ...  Each data point represents one face image. The data points of age from 0 to 93 are colored from blue to red. We choose four nearest neighbors for all the graph embedded learning algorithms.  ... 
doi:10.1109/tmm.2008.921847 fatcat:yhi6bh5wgfeybmn5wnbfh2nwee

Subspace clustering with dense representations

Eva L. Dyer, Christoph Studer, Richard G. Baraniuk
2013 2013 IEEE International Conference on Acoustics, Speech and Signal Processing  
Unions of subspaces have recently been shown to provide a compact nonlinear signal model for collections of highdimensional data, such as large collections of images or videos.  ...  In this paper, we introduce a novel data-driven algorithm for learning unions of subspaces directly from a collection of data; our approach is based upon forming minimum 2 -norm (least-squares) representations  ...  Following this, we showcase the classification performance of DSC for segmenting images of faces under different illumination conditions.  ... 
doi:10.1109/icassp.2013.6638260 dblp:conf/icassp/DyerSB13 fatcat:phjkepjbwrglnowgae3tlod7om

Continuous Manifold Based Adaptation for Evolving Visual Domains

Judy Hoffman, Trevor Darrell, Kate Saenko
2014 2014 IEEE Conference on Computer Vision and Pattern Recognition  
Our approach can learn to distinguish categories using training data collected at some point in the past, and continue to update its model of the categories for some time into the future, without receiving  ...  Adaptation can be achieved via transforms or kernels learned between such stationary source and target subspaces.  ...  This is not what the method was originally designed for, would be very computationally expensive and would require cross-validating or tuning a hyperparameter to choose the appropriate window size.  ... 
doi:10.1109/cvpr.2014.116 dblp:conf/cvpr/HoffmanDS14 fatcat:iw54yhxc3vgbnd5yh2caeyzvpq

Introductory Chapter: Face Recognition - Overview, Dimensionality Reduction, and Evaluation Methods [chapter]

S. Ramakrishnan
2016 Face Recognition - Semisupervised Classification, Subspace Projection and Evaluation Methods  
Other than this introductory chapter, this book has four more chapters, two chapters for dimensionality reduction and one for an overview of the face recognition systems and evaluation methods.  ...  Over the past few decades, researchers in field of computers and electrical and electronics engineering have worked continuously to improve the performances of the face recognition systems.  ...  This will help the students to choose an appropriate technique for doing their projects. Eight different potential applications of face recognition systems are highlighted in Chapter 1.  ... 
doi:10.5772/63995 fatcat:ojk4srxeyrf2nfxkxqvl64x22e

Classification of CASI-3 hyperspectral image by subspace method

Buho Hoshino, Hasi Bagan, Akihiro Nakazawa, Masami Kaneko, Masaki Kawai, Tetuo Yabuki
2011 2011 IEEE International Geoscience and Remote Sensing Symposium  
Then, using the iterative learning technology of averaged learning subspace methods (ALSM) to rotate the subspaces slowly for optimizes the subspaces to get better classification accuracy.  ...  This study presents a supervised subspace learning classification method which can be applied directly to the original set of spectral bands of hyperspectral data for land cover classification purpose.  ...  ACKNOWLEDGMENT This work was supported by Grant-in-Aid for Scientific Research (B) (No. 21370005) from Ministry of Education, Culture, Sports, Science and Technology (MEXT) of Japan and supported by grants  ... 
doi:10.1109/igarss.2011.6049232 dblp:conf/igarss/HoshinoBNKKY11 fatcat:amwlsqojgbao3htocx3n67zozm

Active subspace learning

Xiaofei He, Deng Cai
2009 2009 IEEE 12th International Conference on Computer Vision  
In this paper, we propose a novel active subspace learning algorithm which selects the most informative data points and uses them for learning an optimal subspace.  ...  Experiments on image retrieval show improvement over state-of-the-art methods.  ...  However, most of previous active learning approaches are designed for classification rather than subspace learning.  ... 
doi:10.1109/iccv.2009.5459329 dblp:conf/iccv/HeC09 fatcat:rt56w5gpw5g4vhrzfthayo3bye

Semi-supervised Node Splitting for Random Forest Construction

Xiao Liu, Mingli Song, Dacheng Tao, Zicheng Liu, Luming Zhang, Chun Chen, Jiajun Bu
2013 2013 IEEE Conference on Computer Vision and Pattern Recognition  
The proposed algorithm is compared with state-of-the-art supervised and semi-supervised algorithms for typical computer vision applications such as object categorization and image segmentation.  ...  To avoid the curse of dimensionality, we project the data points from the original high-dimensional feature space onto a low-dimensional subspace before estimation.  ...  Even the abundant data are unlabeled one can still choose the appropriate separating hyperplane as in (c) by combining the law of total probability and the kernel-based density estimation.  ... 
doi:10.1109/cvpr.2013.70 dblp:conf/cvpr/LiuSTLZCB13 fatcat:ywawii4ocredpml2cld6pcxtty

Deep Goal-Oriented Clustering [article]

Yifeng Shi, Christopher M. Bender, Junier B. Oliva, Marc Niethammer
2020 arXiv   pre-print
One could reasonably expect appropriately clustering the data would aid the downstream prediction task and, conversely, a better prediction performance for the downstream task could potentially inform  ...  a more appropriate clustering strategy.  ...  More concretely, if z ∼ p(z|c) for some index c, we assume p(y|z, c) ∝ g c (z) for some subspace-specific g c . As a result, we learn a different mapping function for each subspace.  ... 
arXiv:2006.04259v3 fatcat:hoorwgiohvh7db4aigkuw2e6b4

Neural Collaborative Subspace Clustering [article]

Tong Zhang, Pan Ji, Mehrtash Harandi, Wenbing Huang, Hongdong Li
2019 arXiv   pre-print
We introduce the Neural Collaborative Subspace Clustering, a neural model that discovers clusters of data points drawn from a union of low-dimensional subspaces.  ...  At its heart, our neural model benefits from a classifier which determines whether a pair of points lies on the same subspace or not.  ...  In this paper, we formulate subspace clustering as a binary classification problem through collaborative learning of two modules, one for image classification and the other for subspace affinity learning  ... 
arXiv:1904.10596v1 fatcat:4bt3lvtnhje5faivs6wn5ngkcq

Learning Invariant Features Using Subspace Restricted Boltzmann Machine

Jakub M. Tomczak, Adam Gonczarek
2016 Neural Processing Letters  
We evaluate the behavior of subspaceRBM through experiments with MNIST digit recognition task and Caltech 101 Silhouettes image corpora, measuring cross-entropy reconstruction error and classification  ...  The subspace units reflect variations of a pattern in data and the gate unit is responsible for activating the subspace units. Additionally, the gate unit can be seen as a pooling feature.  ...  Introduction The success of machine learning methods stems from appropriate data representation.  ... 
doi:10.1007/s11063-016-9519-9 fatcat:nfk2nieopng4vaz42dyo6jju6q

Constrained Low-Rank Learning Using Least Squares-Based Regularization

Ping Li, Jun Yu, Meng Wang, Luming Zhang, Deng Cai, Xuelong Li
2017 IEEE Transactions on Cybernetics  
Most low-rank methods are incapable of capturing low-dimensional subspace for supervised learning tasks, e.g., classification and regression.  ...  Low-rank learning has attracted much attention recently due to its efficacy in a rich variety of real-world tasks, e.g., subspace segmentation and image categorization.  ...  ACKNOWLEDGMENTS The authors would like to thank the anonymous reviewers for their helpful and constructive comments that have greatly contributed to improving this manuscript.  ... 
doi:10.1109/tcyb.2016.2623638 pmid:27849552 fatcat:znrgzbzxu5g5pmdcyeb3rmgkgq

Clustering-based subspace SVM ensemble for relevance feedback learning

Rongrong Ji, Hongxun Yao, Jicheng Wang, Pengfei Xu, Xianming Liu
2008 2008 IEEE International Conference on Multimedia and Expo  
This paper presents a subspace SVM ensemble algorithm for adaptive relevance feedback (RF) learning.  ...  Finally, regression results of multiple SVMs are probabilistic assembled to give the final labeling prediction for test image.  ...  Tong [5] proposed a criterion to adaptively select the sampling points that closest to the classification hyper-plane for SVM training in relevance feedback.  ... 
doi:10.1109/icme.2008.4607661 dblp:conf/icmcs/JiYWXL08 fatcat:sy3pw3dwx5bszc6eyebbgn3elu

Ensemble-based hybrid probabilistic sampling for imbalanced data learning in lung nodule CAD

Peng Cao, Jinzhu Yang, Wei Li, Dazhe Zhao, Osmar Zaiane
2014 Computerized Medical Imaging and Graphics  
To solve these challenges, we proposed a hybrid probabilistic sampling combined with diverse random subspace ensemble.  ...  Classification plays a critical role in false positive reduction (FPR) in lung nodule computer aided detection (CAD).  ...  Acknowledgments This work is supported by the Alberta Innovates Centre for Machine Learning as well as the National Natural Science Foundation of China (61001047) and one author was supported by the China  ... 
doi:10.1016/j.compmedimag.2013.12.003 pmid:24418073 fatcat:r6bw75plkzdhvb2v6sikowrn2m

Domain adaptation for object recognition: An unsupervised approach

Raghuraman Gopalan, Ruonan Li, Rama Chellappa
2011 2011 International Conference on Computer Vision  
Motivated by incremental learning, we create intermediate representations of data between the two domains by viewing the generative subspaces (of same dimension) created from these domains as points on  ...  the Grassmann manifold, and sampling points along the geodesic between them to obtain subspaces that provide a meaningful description of the underlying domain shift.  ...  From this we can compute the mean of source subspaces, sayS 1 , and the mean for targetS 2 . A popular method for defining the mean of points on a manifold was proposed by Karcher [24] .  ... 
doi:10.1109/iccv.2011.6126344 dblp:conf/iccv/GopalanLC11 fatcat:o2v5q7svefadhpf3lce5ox462a
« Previous Showing results 1 — 15 out of 11,818 results