20,339 Hits in 3.3 sec

Structure Sparsity for Multi-camera Gait Recognition [chapter]

Qiyue Yin, Rong Sun, Liang Wang, Ran He
2012 Communications in Computer and Information Science  
This paper studies multi-camera gait recognition via structure sparsity.  ...  With the rapid development of surveillance technology, there are often several cameras in one scenario. The multi-camera usage to perform gait recognition becomes a challenge problem.  ...  We also learned an effective dictionary for the proposed structure sparsity method. Experimental results show that our method can efficiently deal with the multi-camera gait recognition problem.  ... 
doi:10.1007/978-3-642-33506-8_33 fatcat:yqj7lfgtijbd7gqsl3d7gpaeie

Multi-view multi-sparsity kernel reconstruction for multi-class image classification

Xiaofeng Zhu, Qing Xie, Yonghua Zhu, Xingyi Liu, Shichao Zhang
2015 Neurocomputing  
This paper addresses the problem of multi-class image classification by proposing a novel multi-view multi-sparsity kernel reconstruction (MMKR for short) model.  ...  Given images (including test images and training images) representing with multiple visual features, the MMKR first maps them into a high-dimensional space, e.g., a reproducing kernel Hilbert space (RKHS  ...  First, multi-view learning employ all the views for each data point without considering the individual characteristics of each data point.  ... 
doi:10.1016/j.neucom.2014.08.106 fatcat:4ouieljmvzddzcdjvu35sfsida

Multi-task Sparse Learning with Beta Process Prior for Action Recognition

Chunfeng Yuan, Weiming Hu, Guodong Tian, Shuang Yang, Haoran Wang
2013 2013 IEEE Conference on Computer Vision and Pattern Recognition  
In this paper, we formulate human action recognition as a novel Multi-Task Sparse Learning(MTSL) framework which aims to construct a test sample with multiple features from as few bases as possible.  ...  In terms of non-informative gamma hyper-priors, the sparsity level is totally decided by the data.  ...  Therefore, we classify j based on how well j is reproduced by the coefficients associated with the learned dictionary of each individual class. For each task j, we de-fineˆj = [θ j,1 , ...  ... 
doi:10.1109/cvpr.2013.61 dblp:conf/cvpr/YuanHTYW13 fatcat:khyvhcfntrbdvnwgf7b4muhava

Multi-View Clustering and Feature Learning via Structured Sparsity

Hua Wang, Feiping Nie, Heng Huang
2013 International Conference on Machine Learning  
In this paper, we propose a novel multi-view learning model to integrate all features and learn the weight for every feature with respect to each cluster individually via new joint structured sparsity-inducing  ...  The proposed multi-view learning framework allows us not only to perform clustering tasks, but also to deal with classification tasks by an extension when the labeling knowledge is available.  ...  multi-view learning methods lies in its capability for simultaneous view selection and individual feature selection.  ... 
dblp:conf/icml/WangNH13a fatcat:iyiyp5ewsfcmxoxcj5mvcp23di

Heterogeneous Visual Features Fusion via Sparse Multimodal Machine

Hua Wang, Feiping Nie, Heng Huang, Chris Ding
2013 2013 IEEE Conference on Computer Vision and Pattern Recognition  
In this paper, We propose a novel Sparse Multimodal Learning (SMML) approach to integrate such heterogeneous features by using the joint structured sparsity regularizations to learn the feature importance  ...  of for the vision tasks from both group-wise and individual point of views.  ...  We will use a new joint structured sparsity regularizers to explore both group-wise and individual importance of each feature for different classes.  ... 
doi:10.1109/cvpr.2013.398 dblp:conf/cvpr/WangNHD13 fatcat:qufsnepjmvagrfb7r5ljwavphi

MNEW: Multi-domain Neighborhood Embedding and Weighting for Sparse Point Clouds Segmentation [article]

Yang Zheng, Izzat H. Izzat, Sanling Song
2020 arXiv   pre-print
We propose a new method called MNEW, including multi-domain neighborhood embedding, and attention weighting based on their geometry distance, feature similarity, and neighborhood sparsity.  ...  The distance/similarity attention and sparsity-adapted weighting mechanism of MNEW enable its capability for a wide range of data sparsity distribution.  ...  However, since we compute adaptive attention & weighting factors in each domain separately, their impact are learned individually.  ... 
arXiv:2004.03401v1 fatcat:yz6k55hntrbypiuys6e3lvvdx4

A Constrained, Weighted-L1 Minimization Approach for Joint Discovery of Heterogeneous Neural Connectivity Graphs [article]

Chandan Singh, Beilun Wang, Yanjun Qi
2017 arXiv   pre-print
To remedy this problem, the paper here introduces a novel, weighted-ℓ_1, multi-task graphical model (W-SIMULE).  ...  Recent studies have used Gaussian graphical models to learn brain connectivity via statistical dependencies across brain regions from neuroimaging.  ...  After preprocessing with this pipeline, 871 individuals remain (468 diagnosed with autism).  ... 
arXiv:1709.04090v2 fatcat:vyera32goventfcynuic5zfkdm

Visual classification with multi-task joint sparse representation

Xiao-Tong Yuan, Shuicheng Yan
2010 2010 IEEE Computer Society Conference on Computer Vision and Pattern Recognition  
A joint sparsity-inducing norm is utilized to enforce class-level joint sparsity patterns among the multiple representation vectors.  ...  We address the problem of visual classification with multiple features and/or multiple instances.  ...  From the viewpoint of multi-task learning, problem ( 5 ) is a multi-task regression model with K × L independent LSR. A.  ... 
doi:10.1109/cvpr.2010.5539967 dblp:conf/cvpr/YuanY10 fatcat:5in5q4wgjnhh7g7wuduq4zfj2u

Adaptive Sparse Representations for Video Anomaly Detection

Xuan Mo, Vishal Monga, Raja Bala, Zhigang Fan
2014 IEEE transactions on circuits and systems for video technology (Print)  
Second, we introduce non-linearity into the linear sparsity model and dictionary design and optimization technique to enable superior class separability and dictionary compactness.  ...  This progress has also been leveraged for sparsity-based video-anomaly detection where test events are expressed as sparse linear combinations of example events from a given (normal or anomalous) class  ...  We also compare our experimental results against another three video anomaly detection methods that can deal with multi-object anomalies: 1.) the sparsity model with online dictionary learning of Zhao  ... 
doi:10.1109/tcsvt.2013.2280061 fatcat:njjuwnbe7fa7toikvl7lsdguhi

Continual Learning via Neural Pruning [article]

Siavash Golkar, Michael Kagan, Kyunghyun Cho
2019 arXiv   pre-print
In order to deal with the possible compromise between model sparsity and performance, we formalize and incorporate the concept of graceful forgetting: the idea that it is preferable to suffer a small amount  ...  We introduce Continual Learning via Neural Pruning (CLNP), a new method aimed at lifelong learning in fixed capacity models based on neuronal model sparsification.  ...  [20] : an MLP with two hidden layers, each with 2000 neurons and ReLU activation and a softmax multi-class cross-entropy loss trained with Adam optimizer and batch size 256.  ... 
arXiv:1903.04476v1 fatcat:tsaehmr2ujav5oxezasjltl5su

Learning a Tree of Metrics with Disjoint Visual Features

Sung Ju Hwang, Kristen Grauman, Fei Sha
2011 Neural Information Processing Systems  
classes (e.g., Persian cat vs.  ...  We validate our approach with multiple image datasets using the WordNet taxonomy, show its advantages over alternative metric learning approaches, and analyze the meaning of attribute features selected  ...  learned with the LMNN algorithm [11] , and 3) Multi-Metric LMNN: one metric learned per class using the multiple metric LMNN variant [11] .  ... 
dblp:conf/nips/HwangGS11 fatcat:xgq2jmgtgbamffso4exw6wa46e

Visual Classification With Multitask Joint Sparse Representation

Xiao-Tong Yuan, Xiaobai Liu, Shuicheng Yan
2012 IEEE Transactions on Image Processing  
A joint sparsity-inducing norm is utilized to enforce class-level joint sparsity patterns among the multiple representation vectors.  ...  We address the problem of visual classification with multiple features and/or multiple instances.  ...  From the viewpoint of multi-task learning, problem (5) is a multi-task regression model with K × L independent LSR. A.  ... 
doi:10.1109/tip.2012.2205006 pmid:22736645 fatcat:lt6lhmwy7zgufgi2bunp2aetfa

Efficient Noisy Sound-Event Mixture Classification Using Adaptive-Sparse Complex-Valued Matrix Factorization and OvsO SVM

Phetcharat Parathai, Naruephorn Tengtrairat, Wai Lok Woo, Mohammed Abdul Muttaleb Abdullah, Gholamreza Rafiee, Ossama Alshabrawy
2020 Sensors  
Mel frequency cepstral coefficients, short-time energy, and short-time zero-crossing rate are learned with multi sound-event classes by the SVM based OvsO method.  ...  The first step of the multi-class MSVM method is to segment the separated signal into blocks by sliding demixed signals, then encoding the three features of each block.  ...  A deep learning framework can be established with two convolutional neural networks (CNNs) and a deep multi-layer perceptron (MLP) with rectified linear units (ReLU) as the activation function [29, 30  ... 
doi:10.3390/s20164368 pmid:32764362 pmcid:PMC7472059 fatcat:3jl5cmk6ezatrjshtwsrogzmaa

Multimodal Task-Driven Dictionary Learning for Image Classification

Soheil Bahrampour, Nasser M. Nasrabadi, Asok Ray, William Kenneth Jenkins
2016 IEEE Transactions on Image Processing  
In this task-driven formulation, the multimodal dictionaries are learned simultaneously with their corresponding classifiers.  ...  In this paper, we propose a multimodal task-driven dictionary learning algorithm under the joint sparsity constraint (prior) to enforce collaborations among multiple homogeneous/heterogeneous sources of  ...  It is consistently observed in all the studied applications that the multimodal dictionary learning algorithm with the mixed prior results in better performance than those with individual 12 or 12 prior  ... 
doi:10.1109/tip.2015.2496275 pmid:26540686 fatcat:xcnlovafdne4bbhzaf7iqvtuxe

An Intelligent Groundwater Management Recommender System

A A Abd El-Aziz, College of Computer and Information Sciences, Jouf University, KSA
2021 Indian Journal of Science and Technology  
Finding: The main goal of our proposed approach is to classify groundwater into multi-labels, which are drinking water (Excellent or Good) or Irrigation water (Poor or Very Poor) with guarantee a higher  ...  Methods : We propose a machine-deep learning model based on a recommender system to manage and classify groundwater.  ...  Applying deep learning methods: We apply the CNN with the Multi-Class Cross-Entropy Loss function. 7.  ... 
doi:10.17485/ijst/v14i37.1332 fatcat:cle2hzvnn5c3tnx4nhddzgioee
« Previous Showing results 1 — 15 out of 20,339 results