520 Hits in 6.4 sec

Sparsity Based Locality-Sensitive Discriminative Dictionary Learning for Video Semantic Analysis

Ben-Bright Benuwa, Yongzhao Zhan, Benjamin Ghansah, Ernest K. Ansah, Andriana Sarkodie
2018 Mathematical Problems in Engineering  
In the proposed algorithm, a discriminant loss function for the category based on sparse coding of the sparse coefficients is introduced into structure of locality-sensitive dictionary learning (LSDL)  ...  Based on the issues stated afore, a novel learning algorithm, called sparsity based locality-sensitive discriminative dictionary learning (SLSDDL) for VSA is proposed in this paper.  ...  More recently, there has been an advancement in sparse dictionary learning for video semantic as proposed in [2] , a video semantic detection method based on locality-sensitive discriminant sparse representation  ... 
doi:10.1155/2018/9312563 fatcat:qnrsbiwnxjdffcj3g5kkcvpxhe

An Efficient Method for Automatic Video Annotation and Retrieval in Visual Sensor Networks

Jiangfan Feng, Wenwen Zhou
2014 International Journal of Distributed Sensor Networks  
We study a general method of discriminative dictionary learning which is independent of the specific dictionary and classifier learning algorithms.  ...  To address the problem of pattern classification in video annotation, this paper proposes a discriminative constraint to find a solution to approach the sparse representative coefficients with discrimination  ...  and Technological Research Program of Chongqing Municipal Education Commission (KJ120526).  ... 
doi:10.1155/2014/832512 fatcat:lqr3p4ca45fp3ijry2l6mc5m5q

Multi-Perspective Cost-Sensitive Context-Aware Multi-Instance Sparse Coding and Its Application to Sensitive Video Recognition

Weiming Hu, Xinmiao Ding, Bing Li, Jianchao Wang, Yan Gao, Fangshi Wang, Stephen Maybank
2016 IEEE transactions on multimedia  
methods outperform the traditional MIL methods and the traditional SVM and KNN-based methods.  ...  Based on color emotion and color harmony theories, we extract visual emotional features from videos.  ...  The results on the violent and horror video datasets show that our methods outperform the traditional MIL-based methods and the traditional SVM and KNN-based methods.  ... 
doi:10.1109/tmm.2015.2496372 fatcat:oeenmbw43jebrivncuixurxwqm

A Generic Framework for Video Annotation via Semi-Supervised Learning

Tianzhu Zhang, Changsheng Xu, Guangyu Zhu, Si Liu, Hanqing Lu
2012 IEEE transactions on multimedia  
In this paper, we propose a novel approach based on semi-supervised learning by means of information from the Internet for interesting event annotation in videos.  ...  However, there lacks a generic framework due to several difficulties, such as dependence of domain knowledge, insufficiency of training data, no precise localization and inefficacy for large-scale video  ...  Therefore, interesting events in a video can be localized and video can be segmented with its semantic concept. 3) Semantic Search and Navigation: Based on event detection (event recognition and localization  ... 
doi:10.1109/tmm.2012.2191944 fatcat:7uaujwzq4nfrto5jaim7bf4ify

Action Disambiguation Analysis Using Normalized Google-Like Distance Correlogram [chapter]

Qianru Sun, Hong Liu
2013 Lecture Notes in Computer Science  
All pairwise distances compose a NGLD correlogram and its normalized form is incorporated into the final action representation.  ...  Recently, Spatial-Temporal Interest Point (STIP) based local features have shown great promise in complex action analysis.  ...  A framework is provided supporting different methods of local feature detection and extraction.  ... 
doi:10.1007/978-3-642-37431-9_33 fatcat:ysbswllgcff4pkxapkwlj4wv7a

Sparse Ensemble Learning for Concept Detection

Sheng Tang, Yan-Tao Zheng, Yu Wang, Tat-Seng Chua
2012 IEEE transactions on multimedia  
This work presents a novel sparse ensemble learning scheme for concept detection in videos.  ...  More importantly, the sparse NMF ensures that an exemplar is projected to only a few bases (localities) with non-zero coefficients.  ...  The BoW representation is a histogram based on a visual vocabulary of 500 visual words clustered by a set of about 500 000 SIFT features [18] , and weighted by an effective soft-weighting scheme for taking  ... 
doi:10.1109/tmm.2011.2168198 fatcat:yg5dvk75qvgilgsxvdrslblguu

A generic framework for event detection in various video domains

Tianzhu Zhang, Changsheng Xu, Guangyu Zhu, Si Liu, Hanqing Lu
2010 Proceedings of the international conference on Multimedia - MM '10  
In this paper, we present a generic event detection approach based on semi-supervised learning and Internet vision.  ...  Concretely, a Graph-based Semi-Supervised Multiple Instance Learning (GSSMIL) algorithm is proposed to jointly explore small-scale expert labeled videos and large-scale unlabeled videos to train the event  ...  The popular methods for the weight assignment include K-Nearest Neighbor (KNN), Gaussian Kernel Similarity (GKS) [5] and Sparsity Induced Similarity measure (SIS) [7] based on sparse decomposition  ... 
doi:10.1145/1873951.1873967 dblp:conf/mm/ZhangXZLL10 fatcat:aznsm4tedzak3nn4ifq22x2p3y

2021 Index IEEE Transactions on Image Processing Vol. 30

2021 IEEE Transactions on Image Processing  
The primary entry includes the coauthors' names, the title of the paper or other item, and its location, specified by the publication abbreviation, year, month, and inclusive pagination.  ...  The Subject Index contains entries describing the item under all appropriate subject headings, plus the first author's name, the publication abbreviation, month, and year, and inclusive pages.  ...  ., +, TIP 2021 3499-3512 A Sparse Representation Based Joint Demosaicing Method for Single-Chip Polarized Color Sensor.  ... 
doi:10.1109/tip.2022.3142569 fatcat:z26yhwuecbgrnb2czhwjlf73qu

Scene Classification Using a Hybrid Generative/Discriminative Approach

A. Bosch, A. Zisserman, X. Muoz
2008 IEEE Transactions on Pattern Analysis and Machine Intelligence  
In this paper we follow the second strategy and introduce a classification algorithm based on a combination of unsupervised probabilistic Latent Semantic Analysis (pLSA) [16] followed by a discriminative  ...  representation for each image, and subsequently training a multi-way classifier on the topic distribution vector for each image.  ...  This is the method proposed by Lazebnik et al. [18] which is based on spatial pyramid matching [13] .  ... 
doi:10.1109/tpami.2007.70716 pmid:18276975 fatcat:fdetcvnxc5fkno2iolwigrdcsi

Unsupervised Feature Learning by Cross-Level Instance-Group Discrimination [article]

Xudong Wang, Ziwei Liu, Stella X. Yu
2021 arXiv   pre-print
To effect both grouping and discrimination objectives, we impose them on features separately derived from a shared representation.  ...  Unsupervised feature learning has made great strides with contrastive learning based on instance discrimination and invariant mapping, as benchmarked on curated class-balanced datasets.  ...  How Sensitive Are Hyper-parameters Weight λ and Temperature T ?  ... 
arXiv:2008.03813v5 fatcat:p4brpjbisrhhxadkynuldnd4bu

Learned features versus engineered features for multimedia indexing

Mateusz Budnik, Efrain-Leonardo Gutierrez-Gomez, Bahjat Safadi, Denis Pellerin, Georges Quénot
2016 Multimedia tools and applications  
In this paper, we compare "traditional" engineered (hand-crafted) features (or descriptors) and learned features for content-based indexing of image or video documents.  ...  Learned (or semantic) features are obtained by training classifiers on a source collection containing samples annotated with concepts.  ...  The authors also wish to thank Florent Perronnin from XRCE for providing features based on classification scores from classifiers trained on ILSVRC/ImageNet data [28] .  ... 
doi:10.1007/s11042-016-4240-2 fatcat:azrhyxhvm5ctxf3xkytwkrtn64

A Comparative Study of Data Mining approaches for Bag of Visual Words Based Image Classification

Sahar Abd El Hameed Mokhtar, Heba Ahmed Elnemr
2018 Journal of Computer Science  
Five data mining techniques; Neural Networks (NN), Decision Trees (DT), Bayesian Network (BN), Discriminant Analysis (DA) and K Nearest Neighbor (KNN), are explored and evaluated on two large different  ...  This paper presents and discusses the application of various data mining techniques for image classification based on Bag of Visual Words (BoVW) feature extraction algorithm.  ...  Acknowledgement This work has been achieved at Computers and Systems Department, Electronics Research Institute.  ... 
doi:10.3844/jcssp.2018.53.66 fatcat:6jtb327ytzbbfi3ybfcwkh4bre

Rethinking Supervised Pre-training for Better Downstream Transferring [article]

Yutong Feng, Jianwen Jiang, Mingqian Tang, Rong Jin, Yue Gao
2022 arXiv   pre-print
To alleviate this problem, we propose a new supervised pre-training method based on Leave-One-Out K-Nearest-Neighbor, or LOOK for short.  ...  Though for most cases, the pre-training stage is conducted based on supervised methods, recent works on self-supervised pre-training have shown powerful transferability and even outperform supervised pre-training  ...  discrimination based methods.  ... 
arXiv:2110.06014v2 fatcat:ainwokqzgvftnfymofufiyzyl4

Spatiotemporal salient points for visual recognition of human actions

A. Oikonomopoulos, I. Patras, M. Pantic
2005 IEEE Transactions on Systems Man and Cybernetics Part B (Cybernetics)  
This paper addresses the problem of human action recognition by introducing a sparse representation of image sequences as a collection of spatiotemporal events that are localized at points that are salient  ...  We propose a classification scheme that is based on Relevance Vector Machines and on the proposed distance measure.  ...  A simple kNN classifier and one based on Relevance Vector Machines, introduced in [18], are used in order to test the efficiency of the representation.We test the proposed method using real image sequences  ... 
doi:10.1109/tsmcb.2005.861864 pmid:16761823 fatcat:afagddgmw5d7bj3gsc6lmqbc3m

Graph Regularized Deep Sparse Representation for Unsupervised Anomaly Detection

Shicheng Li, Shumin Lai, Yan Jiang, Wenle Wang, Yugen Yi, Henry Man Fai Leung
2021 Computational Intelligence and Neuroscience  
Sparse representation (SR) can be regarded as one of matrix factorization (MF) methods, which is a powerful tool for FR. However, there are some limitations in the original SR.  ...  Then, a L1-norm-based sparsity constraint is added to enhance the discriminant ability of the deep features. Finally, a reconstruction error is applied to distinguish anomalies.  ...  Conclusions Different from the traditional MF-based methods, we propose a deep representation framework based on sparse representation named graph regularized deep sparse representation (GRDSR) to learn  ... 
doi:10.1155/2021/4026132 pmid:34777492 pmcid:PMC8580626 fatcat:5hkku6kzb5aazaeydsjpniyezi
« Previous Showing results 1 — 15 out of 520 results