29 Hits in 6.5 sec

Web Image Annotation Via Subspace-Sparsity Collaborated Feature Selection

Zhigang Ma, Feiping Nie, Yi Yang, Jasper R. R. Uijlings, Nicu Sebe
2012 IEEE transactions on multimedia  
To solve the objective function of our method, we propose an efficient iterative algorithm. Extensive experiments are performed on large image databases that are collected from the web.  ...  The number of web images has been explosively growing due to the development of network and storage technology.  ...  In 2009, the Lab for Media Search in National University of Singapore proposed another large scale image database, i.e., NUS-WIDE where all images are from Flickr [28] .  ... 
doi:10.1109/tmm.2012.2187179 fatcat:euw7qr2rmzcshhryl4vqhyfw4m

Fast Multi-label Learning via Hashing [chapter]

Haifeng Hu, Yong Sun, Jiansheng Wu
2015 Lecture Notes in Computer Science  
Particularly, on the dataset NUS-WIDE with 269,648 instances and the dataset Flickr with 565,444 instances where none of existing methods can return results in 24 hours, HashMLL takes only 90 secs and  ...  After that, relied on statistical information attained from all related labels of the neighboring instances, maxi-mum a posteriori (MAP) principle is used to determine the label set for each unseen instance  ...  NUS-WIDE [18] is a web image dataset created by Lab for Media Search in National University of Singapore which is comprising over 269648 images and ground-truth of 81 concepts for the entire dataset.  ... 
doi:10.1007/978-3-319-25159-2_48 fatcat:vjd3ijcyabgnhfx7dxvsa35ag4

Latent semantic sparse hashing for cross-modal similarity search

Jile Zhou, Guiguang Ding, Yuchen Guo
2014 Proceedings of the 37th international ACM SIGIR conference on Research & development in information retrieval - SIGIR '14  
In particular, LSSH uses Sparse Coding to capture the salient structures of images, and Matrix Factorization to learn the latent concepts from text.  ...  Similarity search methods based on hashing for effective and efficient cross-modal retrieval on large-scale multimedia databases with massive text and images have attracted considerable attention.  ...  The NUS-WIDE dataset is a real-word image dataset created by Lab for Media Search in National University of Singapore [28] . This dataset contains 81 concepts but some are scarce.  ... 
doi:10.1145/2600428.2609610 dblp:conf/sigir/ZhouDG14 fatcat:7mkvr5ijyvfxlcc2aryrhtefzm

Cross-Modal Similarity Learning : A Low Rank Bilinear Formulation [article]

Cuicui Kang, Shengcai Liao, Yonghao He, Jian Wang, Wenjia Niu, Shiming Xiang, Chunhong Pan
2015 arXiv   pre-print
Experiments on three well known image-text cross-media retrieval databases show that the proposed method achieves the best performance compared to the state-of-the-art algorithms.  ...  A new approach to the problem has been raised which intends to match features of different modalities directly.  ...  The NUS-WIDE dataset is a large-scale real-world database where the images were crawled from the "Flickr" website by the Media Search Lab in the National University of Singapore [7] .  ... 
arXiv:1411.4738v2 fatcat:vkvnlr4aqbefjii27ng524eyjq

Deep Discrete Prototype Multilabel Learning

Xiaobo Shen, Weiwei Liu, Yong Luo, Yew-Soon Ong, Ivor W. Tsang
2018 Proceedings of the Twenty-Seventh International Joint Conference on Artificial Intelligence  
DBPC compresses the database into a small set of short discrete prototypes, and uses the prototypes for prediction.  ...  The benefit of DBPC comes from two aspects: 1) The number of distance comparisons are reduced in the prototype; 2) The distance computation cost is significantly decreased in the reduced space.  ...  Acknowledgments This work was conducted within the Rolls-Royce@Nanyang Technological University Corporate Lab with support from the National Research Foundation (NRF) Singapore under the Corp Lab@University  ... 
doi:10.24963/ijcai.2018/371 dblp:conf/ijcai/0001LLOT18 fatcat:2au6lanmljbqrcnx63fpjdzfka

Improving Web Image Search by Bag-Based Reranking

Lixin Duan, Wen Li, Ivor Wai-Hung Tsang, Dong Xu
2011 IEEE Transactions on Image Processing  
Comprehensive experiments on the challenging real-world data set NUS-WIDE demonstrate our framework with automatic bag annotation can achieve the best performances compared with existing image reranking  ...  Observing that at least a certain portion of a positive bag is of positive instances while a negative bag might also contain positive instances, we further use a more suitable generalized MI (GMI) setting  ...  A. Experimental Setup We use the challenging real-world NUS-WIDE data set [4] for experiments.  ... 
doi:10.1109/tip.2011.2159227 pmid:21659029 fatcat:lyhfodq74vdntib4pxlxhoidvm

Semi-Relaxation Supervised Hashing for Cross-Modal Retrieval

Peng-Fei Zhang, Chuan-Xiang Li, Meng-Yuan Liu, Liqiang Nie, Xin-Shun Xu
2017 Proceedings of the 2017 ACM on Multimedia Conference - MM '17  
Essentially, given a similarity matrix, most of these methods tackle a discrete optimization problem by separating it into two stages, i.e., first relaxing the binary constraints and finding a solution  ...  At the same time, to tackle the optimization problem, it relaxes a part of binary constraints, instead of all of them, by introducing an intermediate representation variable.  ...  ACKNOWLEDGEMENTS This work was partially supported by National Natural Sci-  ... 
doi:10.1145/3123266.3123320 dblp:conf/mm/ZhangLLNX17 fatcat:j2kkgfuwqreb3aryv5b5cy4ugi

Task-adaptive Asymmetric Deep Cross-modal Hashing [article]

Fengling Li, Tong Wang, Lei Zhu, Zheng Zhang, Xinhua Wang
2022 arXiv   pre-print
The superiority of TA-ADCMH is proved on two standard datasets from many aspects.  ...  However, existing researches equally handle the different tasks of cross-modal retrieval, and simply learn the same couple of hash functions in a symmetric way for them.  ...  NUS-WIDE is a real-world web image database and created by Lab for Media Search in National University of Singapore.  ... 
arXiv:2004.00197v2 fatcat:osykjo375jdy5hkpo6kbhxnbsq

Ensemble manifold regularized sparse low-rank approximation for multiview feature embedding

Lefei Zhang, Qian Zhang, Liangpei Zhang, Dacheng Tao, Xin Huang, Bo Du
2015 Pattern Recognition  
unified and discriminative embedding that is optimal for a given task.  ...  of the concatenated multiview feature matrix.  ...  This paper is supported by the National Natural Science Foundation of China under Grants 61401317, 61471274, 91338202 and 91338111.  ... 
doi:10.1016/j.patcog.2014.12.016 fatcat:lpv6nr5jmjb5ngl6lmlaas43me

A Dictionary induced Least Squares Framework for Multi-view Dimensionality Reduction with Multi-manifold Embeddings

Xiang-Jun Shen, Timothy Apasiba Abeo, Jianping Gou, Qi-rong Mao, Bing Kun Bao, Shuying Li
2018 IET Computer Vision  
Also, on many multi-view datasets of visual recognition and web image annotation, the DLSME method demonstrates more effectiveness than Graph-Laplacian PCA (gLPCA), robust PCA-optimal mean, canonical correlation  ...  In this way, PCA is viewed as a special instance of the authors' proposed dictionary induced least squares framework (DLS).  ...  Acknowledgments This work was funded in part by the National Natural Science Foundation of China (Nos. 61572240, 61502208, 61572503, and 61872424), and NUPTSF (grant no. NY218001)  ... 
doi:10.1049/iet-cvi.2018.5135 fatcat:utb6magf6bh3rb7j3l2j7nayry

Hash Ranking With Weighted Asymmetric Distance for Image Search

Yuan Cao, Heng Qi, Jien Kato, Keqiu Li
2017 IEEE Transactions on Computational Imaging  
To evaluate the proposed algorithms, we conduct a large number of experiments on four well-known datasets, namely, SIFT, CIFAR-10, MNIST and NUS-WIDE.  ...  Image search can be viewed as a problem of large-scale Approximate Nearest Neighbor (ANN) search in image feature space.  ...  NUS-WIDE: The NUS-WIDE dataset is a real-world web image database from National University of Singapore [4] algorithms.  ... 
doi:10.1109/tci.2017.2736980 fatcat:rnqxsrlqu5c6xnlrgcijmj5d3e

Sparse Unsupervised Dimensionality Reduction for Multiple View Data

Yahong Han, Fei Wu, Dacheng Tao, Jian Shao, Yueting Zhuang, Jianmin Jiang
2012 IEEE transactions on circuits and systems for video technology (Print)  
Experiments on a toy benchmark image data set and two real-world Web image data sets demonstrate the effectiveness of the proposed algorithms.  ...  Different kinds of high-dimensional visual features can be extracted from a single image.  ...  We used the Real-World Web Image Database from the National University of Singapore (NUS-WIDE) [32] and the Microsoft Research Asia Internet Multimedia Dataset 2.0 (MSRA-MM 2.0) [36] for the test of  ... 
doi:10.1109/tcsvt.2012.2202075 fatcat:unhuzvtwyveobj7wjzmt2rogcy

Social Anchor-Unit Graph Regularized Tensor Completion for Large-Scale Image Retagging [article]

Jinhui Tang, Xiangbo Shu, Zechao Li, Yu-Gang Jiang, Qi Tian
2018 arXiv   pre-print
Experimental results on a real-world social image database well demonstrate the effectiveness of SUGAR-TC, outperforming several related methods.  ...  Second, a tensor completion based on SUGAR is implemented on the original image-tag-user tensor to refine the tags of the anchor images.  ...  Zheng, “Nus-wide: in ACM CIVR, 2009. a real-world web image database from national university of singapore,”  ... 
arXiv:1804.04397v2 fatcat:fxzkgsdcfbfg5ekutmcmwduyru

Multimedia search reranking

Tao Mei, Yong Rui, Shipeng Li, Qi Tian
2014 ACM Computing Surveys  
Such a problem is challenging because the initial search results often have a great deal of noise.  ...  The explosive growth and widespread accessibility of community contributed media content on the Internet have led to a surge of research activity in multimedia search.  ...  This work was also supported in part by National Science Foundation of China (NSFC) 61128007.  ... 
doi:10.1145/2536798 fatcat:6kmga3jo4fa2tp4354wukfuzja

Image pseudo tag generation with Deep Boltzmann machine anc topic-concept similarity map

Satoru Ishikawa, Jorma Laaksonen, Juha Karhunen
2017 2017 International Joint Conference on Neural Networks (IJCNN)  
NUS-WIDE Dataset The NUS-WIDE dataset has been collected by Chua et al [16] at the National University of Singapore. The dataset consists of 269,648 images with associated tags from Flickr.  ...  It is possible to directly use real-world datasets, such as tagged images uploaded on the world wide web.  ... 
doi:10.1109/ijcnn.2017.7966003 dblp:conf/ijcnn/IshikawaLK17 fatcat:bbaio5lcnvewxorpwzuozkaxee
« Previous Showing results 1 — 15 out of 29 results