Filters








1,069 Hits in 6.8 sec

Probabilistic Matrix Factorization With Semantic And Visual Neighborhoods For Image Tag Completion

Dimitrios Rafailidis
2015 Proceedings of the 5th ACM on International Conference on Multimedia Retrieval - ICMR '15  
We present an image tag completion method, namely PMF-SVN, where the key idea is to exploit images' Semantically and Visually similar Neighborhoods (SVNs) in the learning process of a Probabilistic Matrix  ...  We propose a two-step SVN formation algorithm that can generate an image set with the images being both visually and semantically similar.  ...  thus noisy and incomplete image tags in a probabilistic matrix factorization framework.  ... 
doi:10.1145/2671188.2749333 dblp:conf/mir/Rafailidis15 fatcat:brktsovlpzgqflkcqqre7dl2xe

A Hybrid Probabilistic Model for Unified Collaborative and Content-Based Image Tagging

Ning Zhou, W K Cheung, Guoping Qiu, Xiangyang Xue
2011 IEEE Transactions on Pattern Analysis and Machine Intelligence  
We developed a collaborative filtering method based on non-negative matrix factorization (NMF) for tackling this data sparsity issue.  ...  For images with user provided tags, HPM jointly exploits both the image features and the tags in a unified probability framework to recommend additional tags to label the images.  ...  Algorithm 2 NMF-based lightweight CoIT Input: V -Tag-image association matrix of N training images and M tags v p -Binary representation of test image I p m -Neighborhood size K -Number of tags to be  ... 
doi:10.1109/tpami.2010.204 pmid:21079279 fatcat:jerygw4q7bhlfbldue7vsoyk4i

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

Jiangfan Feng, Wenwen Zhou
2014 International Journal of Distributed Sensor Networks  
Although it has been studied extensively, semantic representation of visual information is not well understood.  ...  Automatic video annotation has become an important issue in visual sensor networks, due to the existence of a semantic gap.  ...  and Technological Research Program of Chongqing Municipal Education Commission (KJ120526).  ... 
doi:10.1155/2014/832512 fatcat:lqr3p4ca45fp3ijry2l6mc5m5q

NMF-KNN: Image Annotation Using Weighted Multi-view Non-negative Matrix Factorization

Mahdi M. Kalayeh, Haroon Idrees, Mubarak Shah
2014 2014 IEEE Conference on Computer Vision and Pattern Recognition  
The recent approaches for image annotation, i.e. the problem of assigning tags to images, have two major drawbacks.  ...  We tested our method on two datasets used for evaluation of image annotation and obtained competitive results.  ...  Government is authorized to reproduce and distribute reprints for Governmental purposes notwithstanding any copyright annotation thereon.  ... 
doi:10.1109/cvpr.2014.31 dblp:conf/cvpr/KalayehIS14 fatcat:ffqre322inca7hnhsxuccsmnsq

Event-Based Media Enrichment Using an Adaptive Probabilistic Hypergraph Model

Xueliang Liu, Meng Wang, Bao-Cai Yin, Benoit Huet, Xuelong Li
2015 IEEE Transactions on Cybernetics  
In this paper, we present a method combining semantic inference and multimodal analysis for automatically finding media content to illustrate events using an adaptive probabilistic hypergraph model.  ...  Nowadays, with the continual development of digital capture technologies and social media services, a vast number of media documents are captured and shared online to help attendees record their experience  ...  Since it is hard to label the relevance of all images in our dataset, for simplicity we set the relevance score of images with an event machine tag to one, and those without machine tags to zero.  ... 
doi:10.1109/tcyb.2014.2374755 pmid:26470061 fatcat:vz7hu5bihjgdtk72bkj2u4yoti

Tag completion based on belief theory and neighbor voting

Amel Znaidia, Hervé Le Borgne, Céline Hudelot
2013 Proceedings of the 3rd ACM conference on International conference on multimedia retrieval - ICMR '13  
We address the problem of tag completion for automatic image annotation.  ...  The method is evaluated in the context of image classification and that of tag suggestion. The database used for visual neighbors search contains 1.2 million images extracted from Flickr.  ...  They represent the image-tag relation by a tag matrix, and search for the optimal tag matrix consistent with both the observed tags and the pairwise visual similarity between images.  ... 
doi:10.1145/2461466.2461476 dblp:conf/mir/ZnaidiaBH13 fatcat:6bz4tqlwcrgnhmixvdbupevkue

Neighborhood-Preserving Hashing for Large-Scale Cross-Modal Search

Botong Wu, Yizhou Wang
2016 Proceedings of the 2016 ACM on Multimedia Conference - MM '16  
We propose a semi-supervised nonlinear probabilistic cross-modal hashing method, namely Neighborhood-Preserving Hashing (NPH), to alleviate the negative effect due to the variation of dimensionality issue  ...  In the literature of cross-modal search, most methods employ linear models to pursue hash codes that preserve data similarity, in terms of Euclidean distance, both within-modal and across-modal.  ...  Whereas, in practice it is very hard or laborious to obtain complete and accurate semantic tags for all the objects in large-scale datasets.  ... 
doi:10.1145/2964284.2967241 dblp:conf/mm/WuW16 fatcat:3a6hzj7b6zfzpmghw6ttrio4oe

Automatic image semantic interpretation using social action and tagging data

Neela Sawant, Jia Li, James Z. Wang
2010 Multimedia tools and applications  
Novel algorithms have been designed to fuse visual features with noisy social labels and behavioral signals.  ...  The plethora of social actions and annotations (tags, comments, ratings) from online media sharing Websites and collaborative games have induced a paradigm shift in the research on image semantic interpretation  ...  For example, the tag-image matrix T I is obtained by collapsing the 3-D matrix U T I along the user dimension.  ... 
doi:10.1007/s11042-010-0650-8 fatcat:kqu6kyess5f3re554jsueuuzem

Automatic image annotation via label transfer in the semantic space

Tiberio Uricchio, Lamberto Ballan, Lorenzo Seidenari, Alberto Del Bimbo
2017 Pattern Recognition  
Our approach works even with the noisy tags of social users, provided that appropriate denoising is performed.  ...  and browse large-scale image collections.  ...  To account for this fact, the proposed tag relevance measurement takes into account both the number of images with tag t in the visual neighborhood of I (namely k t ) and in the entire collection: f T  ... 
doi:10.1016/j.patcog.2017.05.019 fatcat:prf7sbhbjjdr5c4tumwbq5k3lm

An Efficient and Effective Model Based on Mean Positive Examples for Social Image Annotation

Haiyu Song, Jian Yun, Houjie Li, Mingxiao Zheng, Jinxing Yao, Hailin Lv, Anqi Fang
2020 IEEE Access  
To complete tags for an image, the LSR approach needs modeling for the specific image, which needs expensive time cost.  ...  tag (I, J) = exp(−K2 × Dist L2 (T (I), T (J))) (4) where V (I) and V (J) represent the visual feature vectors of image I and image J, respectively, K1 and K2 are transform factors, and T (I) and T (J)  ... 
doi:10.1109/access.2020.3039625 fatcat:v7fa3orrsvegbdvmwn6ecgyu2y

Spectral Hashing With Semantically Consistent Graph for Image Indexing

Peng Li, Meng Wang, Jian Cheng, Changsheng Xu, Hanqing Lu
2013 IEEE transactions on multimedia  
Semantic hashing is a promising way to accelerate similarity search, which designs compact binary codes for a large number of images so that semantically similar images are mapped to close codes.  ...  Meanwhile, our approach, unlike metric learning, can automatically determine the scale factor during the optimization.  ...  Graph Optimization in SH Given an image dataset with samples , is the corresponding -bit binary code vector for image and is an matrix whose th row is .  ... 
doi:10.1109/tmm.2012.2199970 fatcat:eyqkocabgvfejcmihr6ywadcum

Socializing the Semantic Gap

Xirong Li, Tiberio Uricchio, Lamberto Ballan, Marco Bertini, Cees G. M. Snoek, Alberto Del Bimbo
2016 ACM Computing Surveys  
While existing works vary in terms of their targeted tasks and methodology, they rely on the key functionality of tag relevance, i.e. estimating the relevance of a specific tag with respect to the visual  ...  ., image tag assignment, refinement, and tag-based image retrieval is presented.  ...  Jitao Sang for providing the TensorAnalysis results, and Dr. Meng Wang and Dr. Yue Gao for making the Flickr51 dataset available for this survey.  ... 
doi:10.1145/2906152 fatcat:ir6obqydirgihlu54pnwy4clny

Geo-located image analysis using latent representations

M. Cristani, A. Perina, U. Castellani, V. Murino
2008 2008 IEEE Conference on Computer Vision and Pattern Recognition  
Recently, additional textual "tags" can be associated to images, enriching their semantic interpretation beyond the pure visual aspect, and helping to bridge the so-called semantic gap.  ...  In this paper, we present a statistical method for geo-located image categorization, in which categories are formed by clustering geographically proximal images with similar visual appearance.  ...  Probabilistic Latent Semantic Analysis Originally suited for analyzing text corpi, pLSA [7] has been extensively investigated in the modeling of image collections.  ... 
doi:10.1109/cvpr.2008.4587390 dblp:conf/cvpr/CristaniPCM08 fatcat:gbdzir56a5e4horow6zjwuxjs4

Image Tag Refinement by Regularized Latent Dirichlet Allocation [article]

Jingdong Wang, Jiazhen Zhou, Hao Xu, Tao Mei, Xian-Sheng Hua, and Shipeng Li
2013 arXiv   pre-print
Moreover, both the statistics of tags and visual affinities of images in the corpus are explored to help topic modeling.  ...  However, in reality many manually-assigned tags are irrelevant to image content and hence are not reliable for applications. A lot of recent efforts have been conducted to refine image tags.  ...  For example, the neighborhood voting approach [21] is to recommend the tags by exploring the tags of the visually similar images.  ... 
arXiv:1307.7974v1 fatcat:vl3wmkru3zgfrdirq7yghpw32e

Predicting Viewer Affective Comments Based on Image Content in Social Media

Yan-Ying Chen, Tao Chen, Winston H. Hsu, Hong-Yuan Mark Liao, Shih-Fu Chang
2014 Proceedings of International Conference on Multimedia Retrieval - ICMR '14  
For example, given an image tagged with "yummy food," the viewers are likely to comment "delicious" and "hungry," which we refer to as viewer affect concepts (VAC) in this paper.  ...  To the best of our knowledge, this is the first work explicitly distinguishing intended publisher affect concepts and induced viewer affect concepts associated with social visual content, and aiming at  ...  The idea is to use matrix factorization to discover the latent factors of the conditional probability (P (p k |vj) defined in Eq. 1) and use the optimal factor vectors tj, s k for smoothing missing associations  ... 
doi:10.1145/2578726.2578756 dblp:conf/mir/ChenCHLC14 fatcat:rdovz5pekfavhivcib7cd3fyaa
« Previous Showing results 1 — 15 out of 1,069 results