2,776 Hits in 11.6 sec

Confident Kernel Sparse Coding and Dictionary Learning

Babak Hosseini, Barbara Hammer
2018 2018 IEEE International Conference on Data Mining (ICDM)  
In this work, we propose a novel confident K-SRC and dictionary learning algorithm (CKSC) which focuses on the discriminative reconstruction of the data based on its representation in the kernel space.  ...  In recent years, kernel-based sparse coding (K-SRC) has received particular attention due to its efficient representation of nonlinear data structures in the feature space.  ...  Confident Kernel Sparse Coding and Dictionary Learning We propose a novel kernel-based discriminative sparse coding algorithm with the following training framework T rain : min Γ,U Φ(X) − Φ(X)UΓ 2 F +  ... 
doi:10.1109/icdm.2018.00130 dblp:conf/icdm/HosseiniH18 fatcat:qzedybrchfhfpl3ocnz6k5deiy

Dictionaries for image-based recognition

V. M. Patel, Qiang Qiu, R. Chellappa
2013 2013 Information Theory and Applications Workshop (ITA)  
In recent years, Sparse Representation (SR) and Dictionary Learning (DL) have emerged as powerful tools for efficiently processing of image and video data in non-traditional ways.  ...  We will also explore the use of non-linear kernel SR as well as DL methods in many computer vision problems including object recognition, multimodal biometrics recognition, and domain adaptation.  ...  Furthermore, through the use of Mercer kernels, we showed how sparse representation and dictionary learning methods can be made non-linear.  ... 
doi:10.1109/ita.2013.6502927 dblp:conf/ita/PatelQC13 fatcat:zavh6x6da5bthmx3reyjczhlcq

Decoupling Sparse Coding with Fusion of Fisher Vectors and Scalable SVMs for Large-Scale Visual Recognition

Zhengping Ji
2013 2013 IEEE Conference on Computer Vision and Pattern Recognition Workshops  
We also investigate the performance of sparse coding by comparing different combination of algorithms in learning a dictionary and sparse representations.  ...  Although there is a natural pair of algorithms to learn a dictionary and sparse representations (e.g., K-SVD with respect to Orthogonal Matching Pursuit), breaking such a pair and rematching are found  ...  Brumby and other DSGM members in Los Alamos National Laboratory for their helpful discussion and comments.  ... 
doi:10.1109/cvprw.2013.74 dblp:conf/cvpr/Ji13 fatcat:bd2quejuyzcidizlw23qr25lm4

Ambiguously Labeled Learning Using Dictionaries

Yi-Chen Chen, Vishal M. Patel, Rama Chellappa, P. Jonathon Phillips
2014 IEEE Transactions on Information Forensics and Security  
At each iteration of the algorithm, two alternating steps are performed: 1) a confidence update and 2) a dictionary update.  ...  Furthermore, using the kernel methods, we make the dictionary learning framework nonlinear based on the soft decision rule.  ...  Similar to the the linear K-SVD [24] algorithm, the optimization of (15) involves sparse coding and dictionary update steps in the feature space which results in the kernel K-SVD algorithm [29] .  ... 
doi:10.1109/tifs.2014.2359642 fatcat:mlrt6h4gm5hhlipwafvledcb7a

Learning Convolutional Feature Hierarchies for Visual Recognition

Koray Kavukcuoglu, Pierre Sermanet, Y-Lan Boureau, Karol Gregor, Michaël Mathieu, Yann LeCun
2010 Neural Information Processing Systems  
While sparse coding has become an increasingly popular method for learning visual features, it is most often trained at the patch level.  ...  We propose an unsupervised method for learning multi-stage hierarchies of sparse convolutional features.  ...  learned with patch based sparse coding model.  ... 
dblp:conf/nips/KavukcuogluSBGML10 fatcat:pqyyn7csw5febp3mnwjsrfsnau

IBM Research at Image CLEF 2015: Medical Clustering Task

Suman Sedai, Xi Liang, Mani Abedini, Qiang Chen, Rajib Chakravorty, Rahil Garnavi
2015 Conference and Labs of the Evaluation Forum  
The key components used in our submissions are based on sparse coding of SIFT, local binary patterns and multi-scale local binary patterns with spatial pyramid, advanced fisher vector, various SVM kernels  ...  In this paper, we present the learning strategies and feature extraction techniques that were applied by the IBM Research Australia team to the Medical Clustering challenge of ImageCLEF 2015.  ...  Rather than using pre-defined dictionaries, sparse coding algorithms aim to learn a dictionary of basis functions.  ... 
dblp:conf/clef/SedaiLACCG15 fatcat:7ou44d4u6fhchjrvls5yrr5ry4

Learning Bimodal Structure in Audio–Visual Data

G. Monaci, P. Vandergheynst, F.T. Sommer
2009 IEEE Transactions on Neural Networks  
The proposed algorithm uses unsupervised learning to form dictionaries of bimodal kernels from audio-visual material.  ...  A novel model is presented to learn bimodally informative structures from audio-visual signals. The signal is represented as a sparse sum of audio-visual kernels.  ...  Monaci and by NSF through the grant n • 1-11863-26696-44-EUFTS to F. T. Sommer.  ... 
doi:10.1109/tnn.2009.2032182 pmid:19963447 fatcat:mwkyneuugbgkvp5vaxlo7j2ew4

Generalized Pooling for Robust Object Tracking

Bo Ma, Hongwei Hu, Jianbing Shen, Yangbiao Liu, Ling Shao
2016 IEEE Transactions on Image Processing  
Index Terms-Object tracking, feature pooling, Fisher kernel, local coordinate coding, sparse coding. 1057-7149  ...  The high-order statistics and the correlations between responses to different dictionary items are neglected.  ...  is its corresponding sparse code and D p is the dictionary learned with partial representations of target templates.  ... 
doi:10.1109/tip.2016.2588329 pmid:27392358 fatcat:7vmdsq6d2zbtrfxqk2m7q3riqe

Denoising Dictionary Learning Against Adversarial Perturbations [article]

John Mitro and Derek Bridge and Steven Prestwich
2018 arXiv   pre-print
For each model we recorded its accuracy both on the perturbed test data previously misclassified with high confidence and on the denoised one after the reconstruction using dictionary learning.  ...  We examined denoising dictionary learning on MNIST and CIFAR10 perturbed under two different perturbation techniques, fast gradient sign (FGSM) and jacobian saliency maps (JSMA).  ...  Another advantage of dictionary learning and sparse coding is the fact that it can be embedded in any supervised learning algorithm without any severe restrictions.  ... 
arXiv:1801.02257v1 fatcat:cspn3zr6erflbjwu5sp6wt7ebe

Local structure preserving sparse coding for infrared target recognition

Jing Han, Jiang Yue, Yi Zhang, Lianfa Bai, Zhao Zhang
2017 PLoS ONE  
enough local sparse structures to learn a sufficient sparse structure dictionary of a target class.  ...  Sparse coding performs well in image classification. However, robust target recognition requires a lot of comprehensive template images and the sparse learning process is complex.  ...  All these subitems are calculable, so the efficient sparse coding algorithm can also be employed to get KLSPSc coefficientsŜ. Second, learn the dictionary basis B and B ϕ .  ... 
doi:10.1371/journal.pone.0173613 pmid:28323824 pmcid:PMC5360252 fatcat:fdl5vmflgvckbowzfzoysuwamu

Dictionary Learning with Accumulator Neurons [article]

Gavin Parpart, Carlos Gonzalez, Terrence C. Stewart, Edward Kim, Jocelyn Rego, Andrew O'Brien, Steven Nesbit, Garrett T. Kenyon, Yijing Watkins
2022 arXiv   pre-print
Non-spiking LCA has previously been used to achieve unsupervised learning of spatiotemporal dictionaries composed of convolutional kernels from raw, unlabeled video.  ...  We demonstrate how unsupervised dictionary learning with spiking LCA (S-LCA) can be efficiently implemented using accumulator neurons, which combine a conventional leaky-integrate-and-fire (LIF) spike  ...  Unsupervised dictionary learning via sparse coding accounts for many aspects of cortical development.  ... 
arXiv:2205.15386v1 fatcat:issiovcjxvbnbkxomxatjonf7m

Single-Image Super-Resolution via Adaptive Joint Kernel Regression

Chen Huang, Xiaoqing Ding, Chi Fang
2013 Procedings of the British Machine Vision Conference 2013  
Adaptive dictionary learning and dictionary-based sparsity prior are also introduced to interact with the regression prior for robustness.  ...  We further propose a measure called regional redundancy to determine the confidence of these regression groups and thus control their relative effects of regularization adaptively.  ...  [23] , and to two dictionary-based methods, Centralized Sparse Representation (CSR) [5] and Sparse Coding (SC) method [21] .  ... 
doi:10.5244/c.27.101 dblp:conf/bmvc/HuangDF13 fatcat:hl4vmpc74vew3cnjcphaiw75wm

Visual tracking by dictionary learning and motion estimation

Amin Jourabloo, Behnam Babagholami-Mohamadabadi, Amir H. Feghahati, Mohammad T. Manzuri-Shalmani, Mansour Jamzad
2012 2012 IEEE International Symposium on Signal Processing and Information Technology (ISSPIT)  
Here, we utilize motion information in order to reduce this computation time by not calculating sparse codes for all the frames.  ...  The proposed method combines sparse representation and motion estimation to track an object. Recently, sparse representation has gained much attention in signal processing and computer vision.  ...  The proposed method consists of three main steps, exact sparse coding, approximate sparse coding and, dictionary learning which are going as follows. A.  ... 
doi:10.1109/isspit.2012.6621300 dblp:conf/isspit/JourablooBFSJ12 fatcat:ukegzy65pnggph2zt7wuczvpjq

Enhancing Action Recognition by Cross-Domain Dictionary Learning

Fan Zhu, Ling Shao
2013 Procedings of the British Machine Vision Conference 2013  
through which a reconstructive, discriminative and domain-adaptive dictionary-pair can be learned.  ...  The data distribution of relevant actions from a source dataset is adapted to match the data distribution of actions in the target dataset via a cross-domain discriminative dictionary learning method,  ...  Recently, dictionary learning for sparse representation has attracted much attention.  ... 
doi:10.5244/c.27.52 dblp:conf/bmvc/ZhuS13 fatcat:p3cg4bbnfbdpro3bgkgpygfccu

Non-linear dictionary learning with partially labeled data

Ashish Shrivastava, Vishal M. Patel, Rama Chellappa
2015 Pattern Recognition  
Using the kernel method, we propose a non-linear discriminative dictionary learning technique which utilizes both labeled and unlabeled data for learning dictionaries in the high-dimensional feature space  ...  While recent techniques for discriminative dictionary learning have demonstrated tremendous success in image analysis applications, their performance is often limited by the amount of labeled data available  ...  Sparse representation and dictionary learning methods for unsupervised learning have also been proposed.  ... 
doi:10.1016/j.patcog.2014.07.031 fatcat:xqdwg7i5jngnjafc22xwyohg2q
« Previous Showing results 1 — 15 out of 2,776 results