19,806 Hits in 4.2 sec

Graph mode-based contextual kernels for robust SVM tracking

Xi Li, Anthony Dick, Hanzi Wang, Chunhua Shen, Anton van den Hengel
2011 2011 International Conference on Computer Vision  
Finally, this contextual kernel is embedded into SVMs for robust tracking. Experimental results on challenging videos demonstrate the effectiveness and robustness of the proposed tracker.  ...  In this paper, we propose a visual tracker based on support vector machines (SVMs), for which a novel graph mode-based contextual kernel is designed to effectively capture the higher-order contextual information  ...  V 1 V 2 V 3 V 4 V 5 V Conclusion We have proposed a graph mode-based contextual kernel for robust SVM tracking.  ... 
doi:10.1109/iccv.2011.6126364 dblp:conf/iccv/LiDWSH11 fatcat:7irp5yeg7jhujoetzjus3ptn7e

Adaptive and Scalable Android Malware Detection through Online Learning [article]

Annamalai Narayanan, Liu Yang, Lihui Chen, Liu Jinliang
2016 arXiv   pre-print
In a large-scale comparative analysis with more than 87,000 apps, DroidOL achieves 84.29% accuracy outperforming two state-of-the-art malware techniques by more than 20% in their typical batch learning  ...  In order to perform accurate detection, security-sensitive behaviors are captured from apps in the form of inter-procedural control-flow sub-graph features using a state-of-the-art graph kernel.  ...  For SVM, we experiment with the following variants: SVM-Once, SVM-Daily, SVM-MultiOnce, and SVM-MultiDaily. For SVM-Once, SVM classifier is trained only once on the apps from Day 1 (1 Jan '14) .  ... 
arXiv:1606.07150v2 fatcat:klf5uaukzreidojvdx53egs57e

Advances in Hyperspectral Image Classification: Earth Monitoring with Statistical Learning Methods

Gustavo Camps-Valls, Devis Tuia, Lorenzo Bruzzone, Jon Atli Benediktsson
2014 IEEE Signal Processing Magazine  
New methods have been presented to account for the spatial homogeneity of images, to include user's interaction via active learning, to take advantage of the manifold structure with semisupervised learning  ...  , to extract and encode invariances, or to adapt classifiers and image representations to unseen yet similar scenes.  ...  This makes the adaptation process more robust than in the case of semi-supervised learning at the cost of requiring additional labeled samples.  ... 
doi:10.1109/msp.2013.2279179 fatcat:zocewzfdkbetfk4m3ocved7rpy

Generalized Learning Graph Quantization [chapter]

Brijnesh J. Jain, Klaus Obermayer
2011 Lecture Notes in Computer Science  
This contribution extends generalized LVQ, generalized relevance LVQ, and robust soft LVQ to the graph domain.  ...  The proposed approaches are based on the basic learning graph quantization (lgq) algorithm using the orbifold framework.  ...  We compared the lgq algorithms with the similarity kernel in conjunction with the SVM (sk+svm) and the family of Lipschitz embeddings in conjunction with SVM (ls+svm) proposed by [15] .  ... 
doi:10.1007/978-3-642-20844-7_13 fatcat:js5f4gooebh3hishc3okvdpocm

Achievement of Small Target Detection for Sea Ship Based on CFAR-DBN

LiBing Yuan, XueLi Chi, Hui Wei, Zhiguo Qu
2022 Wireless Communications and Mobile Computing  
Secondly, the output data of the hidden layer of the last layer of DBN are used as the input data of SVM, and the trained DBN model is applied to local detection, so as to improve the accuracy and robustness  ...  With the rapid development of marine exploration and marine transportation, the activities of marine ships are becoming more and more frequent.  ...  [14] proposed a CFAR with adaptive selection of pixel sets, which performs well in uniform clutter environments.  ... 
doi:10.1155/2022/4630155 fatcat:affvvtk2zbcv5p4pjnokkzpcxu

Classifying Limited Resource Data Using Semi-supervised SVM

Naralasetti Veeranjaneyulu, Jyostna Bodapati, Suvarna Buradagunta
2020 Ingénierie des Systèmes d'Information  
Semi-supervised learning is an approach that uses additionally available unlabeled data to improve performance of a model with limited data.  ...  To gain better performance with limited training data we suggest semisupervised SVM models for EEG Signal classification and image classification tasks.  ...  Though any other ML algorithm can be adapted to semi-supervised setting, we prefer to use SVM as it is robust with small scale data.  ... 
doi:10.18280/isi.250315 fatcat:nxhm4ygsebacdi6j3kuvcu2fbi

Group Tracking: Exploring Mutual Relations for Multiple Object Tracking [chapter]

Genquan Duan, Haizhou Ai, Song Cao, Shihong Lao
2012 Lecture Notes in Computer Science  
We update relational graphs through analyzing object trajectories and cast the relational weight learning task as an online latent SVM problem.  ...  Instead of considering objects individually, we model objects in mutual context with each other to benefit robust and accurate tracking.  ...  Our approaches, with fixed relational graphs and the adaptive relational graph, all show significantly improvements than MIL on the man, and particularly the adaptive relational graph improves the detection  ... 
doi:10.1007/978-3-642-33712-3_10 fatcat:gg72rtpb7rh2zjp6d5wadciop4

Domain adaptation from multiple sources via auxiliary classifiers

Lixin Duan, Ivor W. Tsang, Dong Xu, Tat-Seng Chua
2009 Proceedings of the 26th Annual International Conference on Machine Learning - ICML '09  
We propose a multiple source domain adaptation method, referred to as Domain Adaptation Machine (DAM), to learn a robust decision function (referred to as target classifier) for label prediction of patterns  ...  We introduce a new datadependent regularizer based on smoothness assumption into Least-Squares SVM (LS-SVM), which enforces that the target classifier shares similar decision values with the auxiliary  ...  Recently, several domain adaptation methods (Crammer et al., 2008; Luo et al., 2008; Mansour et al., 2009) were proposed to learn robust classifiers with the diverse training data from multiple source  ... 
doi:10.1145/1553374.1553411 dblp:conf/icml/DuanTXC09 fatcat:nrtgo4ekkvdaboylflbhvrwwcu

Label Noise Cleansing with Sparse Graph for Hyperspectral Image Classification

Qingming Leng, Haiou Yang, Junjun Jiang
2019 Remote Sensing  
In this paper, we propose a spectral–spatial sparse graph-based adaptive label propagation (SALP) algorithm to address a more practical case, where the label information is contaminated by random noise  ...  adjusted along with its spatial position in the corresponding homogeneous region.  ...  RF that pertain to the label-robust learning paradigm and can be less influenced by noisy labels, NN and SVM are more sensitive to label noise.  ... 
doi:10.3390/rs11091116 fatcat:vc64wkvjija2ddnbbukqn5bwde

Detection of texts in natural images [article]

Gowtham Rangarajan Raman
2014 arXiv   pre-print
Finally a Gabor Feature based SVM is used to classify the presence of text in the candidates. The proposed method was tested with ICDAR 10 dataset and few other images available on the internet.  ...  The image is preprocessed and and edge graph is calculated using a probabilistic framework to compensate for photometric noise.  ...  Hence a lot of importance is placed to ensure that the graph generated captures the details and remains robust during poor lighting.  ... 
arXiv:1411.0126v1 fatcat:siihwtet6zc2dmjegvohu4jmxa

RGB-T Object Tracking:Benchmark and Baseline [article]

Chenglong Li, Xinyan Liang, Yijuan Lu, Nan Zhao, Jin Tang
2018 arXiv   pre-print
approach to learn a robust object representation for RGB-T tracking.  ...  In particular, the tracked object is represented with a graph with image patches as nodes.  ...  S-SVM to achieve robust RGB-T tracking.  ... 
arXiv:1805.08982v1 fatcat:jxkhfxbelnfbli7sfzynvwh5cm

Discriminative Local Sparse Representations for Robust Face Recognition [article]

Yi Chen, Umamahesh Srinivas, Thong T. Do, Vishal Monga, Trac D. Tran
2011 arXiv   pre-print
In particular, we learn discriminative graphs on sparse representations obtained from distinct local slices of a face.  ...  In this paper, we propose a simple yet robust local block-based sparsity model, using adaptively-constructed dictionaries from local features in the training data, to overcome this misalignment problem  ...  However, learning graphs with arbitrarily complex structure is known to be an NP-hard problem [30] .  ... 
arXiv:1111.1947v1 fatcat:jnimvibzuram5eialjmdemikti

Noise-adjusted sparsity-preserving-based dimensionality reduction for hyperspectral image classification

Nam Ly, Qian Du, James E. Fowler
2012 7th IAPR Workshop on Pattern Recognition in Remote Sensing (PRRS)  
In conjunction with the state-of-the-art hyperspectral image classifier, support vector machine with composite kernels (SVM-CK), the experimental study show that NASP can significantly improve the classification  ...  In this paper, we investigate the performance of a sparsity-preserving graph embedding based approach, called [I graph, in hyperspectral image dimensionality reduction (DR), and propose noise-adjusted  ...  II-graph is therefore a non-parametric method. (3) datum-adaptive neighborhood: the number of neighbors defined by 11_ graph is adaptive to each sample, which is valuable for aplications with unevenly  ... 
doi:10.1109/pprs.2012.6398318 fatcat:5igk3y22kjewtdxzye4kcdpcde

Predictive Distribution Matching SVM for Multi-domain Learning [chapter]

Chun-Wei Seah, Ivor W. Tsang, Yew-Soon Ong, Kee-Khoon Lee
2010 Lecture Notes in Computer Science  
Taking this cue, the Predictive Distribution Matching SVM (PDM-SVM) is proposed to learn a robust classifier in the target domain (referred to as the target classifier) by leveraging the labeled data from  ...  In particular, a k-nearest neighbor graph is iteratively constructed to identify the regions of relevant source labeled data where the predictive distribution maximally aligns with that of the target data  ...  Acknowledgement This research was supported by Singapore NTU AcRF Tier-1 Research Grant (RG15/08) and NTU and IHPC joint project entitled "Large Scale Domain Adaptation Machines: Information Integration  ... 
doi:10.1007/978-3-642-15880-3_21 fatcat:gs55ty53cjdmdghdi3vibr7zae

Robust tracking via weakly supervised ranking SVM

Yancheng Bai, Ming Tang
2012 2012 IEEE Conference on Computer Vision and Pattern Recognition  
from frame t + 1 to adapt to substantial changes of the appearance.  ...  To copy with this problem, in this paper we propose a novel algorithm -online Laplacian ranking support vector tracker (LRSVT) -to robustly locate the object.  ...  SVM (Problem 7 and Eq.(8)), -else -Learn the ranking function F (x) with γ I = 0, (Problem 7 with γ I = 0), [ 3 ] 3 Y.  ... 
doi:10.1109/cvpr.2012.6247884 dblp:conf/cvpr/BaiT12 fatcat:uojmuwsnrje2jlwabwcglhq4ge
« Previous Showing results 1 — 15 out of 19,806 results