Filters








97 Hits in 5.2 sec

Special issue on visual information retrieval

Michael S. Lew
2016 International Journal of Multimedia Information Retrieval  
With the flood of images and video from diverse sources (e.g., smartphones, NetFlix, FLICKR, Amazon, Instagram, Twitter, etc.), there is a great need to be able to browse, index and search through the  ...  This special issue provides an overview of important work and the leading research directions in the area of visual information retrieval.  ...  In the paper, "User-adaptive image retrieval via fusing pointwise and pairwise labels" by Lin Chen, Peng Zhang and Baoxin Li, the authors propose a hybrid ranking framework which results in significant  ... 
doi:10.1007/s13735-016-0094-7 fatcat:x437dbgd35fupnhjprfjlei5mu

IRGAN

Jun Wang, Lantao Yu, Weinan Zhang, Yu Gong, Yinghui Xu, Benyou Wang, Peng Zhang, Dell Zhang
2017 Proceedings of the 40th International ACM SIGIR Conference on Research and Development in Information Retrieval - SIGIR '17  
This paper provides a unified account of two schools of thinking in information retrieval modelling: the generative retrieval focusing on predicting relevant documents given a query, and the discriminative  ...  On one hand, the discriminative model, aiming to mine signals from labelled and unlabelled data, provides guidance to train the generative model towards fitting the underlying relevance distribution over  ...  In this paper, we provide an alternative way of training and fusing retrieval models.  ... 
doi:10.1145/3077136.3080786 dblp:conf/sigir/WangYZGXWZZ17 fatcat:l3luiqya4zeunebhwzg6mrxme4

Unsupervised t-Distributed Video Hashing and Its Deep Hashing Extension

Yanbin Hao, Tingting Mu, John Y. Goulermas, Jianguo Jiang, Richang Hong, Meng Wang
2017 IEEE Transactions on Image Processing  
By adapting the corresponding optimization objective and constructing the hash mapping function via a deep neural network, we develop a robust unsupervised training strategy for a deep hashing network.  ...  over pairwise distances, and also a discrete neighbor score based on the cardinality of reciprocal neighbors.  ...  ACKNOWLEDGEMENTS This work was jointly supported by the "China Scholarship Council" and the "International Research Base for Developing Innovative Gerontechnology" sponsored by the national "111" Project  ... 
doi:10.1109/tip.2017.2737329 pmid:28796619 fatcat:dp27jriy5ba27hp6gthykjebeu

Improving Pairwise Ranking for Multi-label Image Classification [article]

Yuncheng Li, Yale Song, Jiebo Luo
2017 arXiv   pre-print
In this work, we propose two techniques to improve pairwise ranking based multi-label image classification: (1) we propose a novel loss function for pairwise ranking, which is smooth everywhere and thus  ...  Pairwise ranking, in particular, has been successful in multi-label image classification, achieving state-of-the-art results on various benchmarks.  ...  would like to explore ways to leverage the distinctive properties of multi-label problem, such as label dependency, label sparsity, and missing labels.  ... 
arXiv:1704.03135v3 fatcat:gfg7enxv6rhk7fk4wl2vlxqvwu

Improving Pairwise Ranking for Multi-label Image Classification

Yuncheng Li, Yale Song, Jiebo Luo
2017 2017 IEEE Conference on Computer Vision and Pattern Recognition (CVPR)  
In this work, we propose two techniques to improve pairwise ranking based multi-label image classification: (1) we propose a novel loss function for pairwise ranking, which is smooth everywhere and thus  ...  Pairwise ranking, in particular, has been successful in multi-label image classification, achieving state-of-the-art results on various benchmarks.  ...  [30] fused the pointwise and pairwise labeling to jointly improve the annotation and attribute prediction accuracy.  ... 
doi:10.1109/cvpr.2017.199 dblp:conf/cvpr/LiSL17 fatcat:dmveosvvcvfttke53bxfeexzhq

Content-Based Features to Rank Influential Hidden Services of the Tor Darknet [article]

Mhd Wesam Al-Nabki, Eduardo Fidalgo, Enrique Alegre, Deisy Chaves
2019 arXiv   pre-print
The unevenness importance of criminal activities in the onion domains of the Tor Darknet and the different levels of their appeal to the end-user make them tangled to measure their influence.  ...  Our approach comprises a modeling unit that represents an onion domain using forty features extracted from five different resources: user-visible text, HTML markup, Named Entities, network topology, and  ...  • To select the best LtR schema, we explore and compare three popular architectures: pointwise, pairwise, and listwise. • Finally, we create a manually ranked dataset, that plays the role of ground  ... 
arXiv:1910.02332v1 fatcat:wzx5u5crvrhbfgisnpnqlibz3q

Data-Driven Shape Analysis and Processing

Kai Xu, Vladimir G. Kim, Qixing Huang, Evangelos Kalogerakis
2016 Computer graphics forum (Print)  
, modelling and exploration, as well as scene analysis and synthesis.  ...  We conclude our report with ideas that can inspire future research in data-driven shape analysis and processing.  ...  Kai Xu is supported by NSFC (61572507, 61202333 and 61532003).  ... 
doi:10.1111/cgf.12790 fatcat:q76sq2syjvce5fjjjb7yoafww4

Combining latent factor model with location features for event-based group recommendation

Wei Zhang, Jianyong Wang, Wei Feng
2013 Proceedings of the 19th ACM SIGKDD international conference on Knowledge discovery and data mining - KDD '13  
of groups and users.  ...  Recommending groups that users are really interested to join is significant for both users and social media.  ...  However, we cannot directly adapt it to our problem because tags are separately labeled by groups and users for themselves, which leads to two pairwise relations instead of one ternary relation.  ... 
doi:10.1145/2487575.2487646 dblp:conf/kdd/ZhangWF13 fatcat:aafhdtqalvf6jgr3b3xdnzebei

From Visual Attributes to Adjectives through Decompositional Distributional Semantics

Angeliki Lazaridou, Georgiana Dinu, Adam Liska, Marco Baroni
2015 Transactions of the Association for Computational Linguistics  
We can thus achieve better attribute (and object) label retrieval by treating images as "visual phrases", and decomposing their linguistic representation into an attribute-denoting adjective and an object-denoting  ...  to nominal labels can implicitly learn to map attributes to adjectives.  ...  , 1936) in several vision-and-language tasks, such as image and caption retrieval (Gong et al., 2014; Hardoon et al., 2004; Hodosh et al., 2013) , we adapt normalized Canonical Correlations Analysis  ... 
doi:10.1162/tacl_a_00132 fatcat:cig3svaf75f57jjgo2r4bxfeum

From Visual Attributes to Adjectives through Decompositional Distributional Semantics [article]

Angeliki Lazaridou, Georgiana Dinu, Adam Liska, Marco Baroni
2015 arXiv   pre-print
We can thus achieve better attribute (and object) label retrieval by treating images as "visual phrases", and decomposing their linguistic representation into an attribute-denoting adjective and an object-denoting  ...  to nominal labels can implicitly learn to map attributes to adjectives.  ...  , 1936) in several vision-and-language tasks, such as image and caption retrieval (Gong et al., 2014; Hardoon et al., 2004; Hodosh et al., 2013) , we adapt normalized Canonical Correlations Analysis  ... 
arXiv:1501.02714v2 fatcat:n4bngfxb7bdwbko65v66lkd2cu

Neural Ranking Models for Document Retrieval [article]

Mohamed Trabelsi, Zhiyu Chen, Brian D. Davison, Jeff Heflin
2021 arXiv   pre-print
We also show the analogy between document retrieval and other retrieval tasks where the items to be ranked are structured documents, answers, images and videos.  ...  Recently, researchers have leveraged deep learning models in information retrieval.  ...  Recent advance in content-based image retrieval: A literature survey. arXiv: 1706. 06064.  ... 
arXiv:2102.11903v1 fatcat:zc2otf456rc2hj6b6wpcaaslsa

Collaborative Memory Network for Recommendation Systems

Travis Ebesu, Bin Shen, Yi Fang
2018 The 41st International ACM SIGIR Conference on Research & Development in Information Retrieval - SIGIR '18  
Motivated by the success of Memory Networks, we fuse a memory component and neural attention mechanism as the neighborhood component.  ...  The associative addressing scheme with the user and item memories in the memory module encodes complex user-item relations coupled with the neural attention mechanism to learn a user-item specific neighborhood  ...  The traditional pointwise approach assumes the user is not interested in the item i − but in reality may not be aware of the item.  ... 
doi:10.1145/3209978.3209991 dblp:conf/sigir/EbesuSF18 fatcat:dixbrvd6o5gd5kqziruxyn2hb4

Deep Learning based Recommender System: A Survey and New Perspectives [article]

Shuai Zhang, Lina Yao, Aixin Sun, Yi Tay
2018 arXiv   pre-print
The influence of deep learning is also pervasive, recently demonstrating its effectiveness when applied to information retrieval and recommender systems research.  ...  Finally, we expand on current trends and provide new perspectives pertaining to this new exciting development of the field.  ...  It compares two images (one positive image user likes and one negative image user dislikes) against a user. e training data is made up of triplets: t (user U t , positive image I + t , negative image I  ... 
arXiv:1707.07435v6 fatcat:2q2dbfy2jvdydhbrmmbyrzctnq

CoupleNet: Paying Attention to Couples with Coupled Attention for Relationship Recommendation [article]

Yi Tay, Anh Tuan Luu, Siu Cheung Hui
2018 arXiv   pre-print
CoupleNet utilizes hierarchical recurrent neural models for learning representations of user profiles and subsequently coupled attention mechanisms to fuse information aggregated from two users.  ...  Intuitively, our approach performs both user profiling and match-making within a unified end-to-end framework.  ...  There are mainly three types of learning to rank methods, namely pointwise, pairwise and list-wise.  ... 
arXiv:1805.11535v1 fatcat:yyctnvrkrvewnc7km2i7tdegdi

Neural Networks in Big Data and Web Search

Will Serrano
2018 Data  
The essential user experience is the self-assurance that the results provided are relevant and exhaustive.  ...  In addition to the challenge of crawling and indexing information within the enormous size and scale of the Internet, e-commerce customers and general Web users should not stay confident that the products  ...  , including matching, translation, classification, structured prediction, search, question answering, and image retrieval.  ... 
doi:10.3390/data4010007 fatcat:2irxpdvtfrclrbndkrubl5jvqq
« Previous Showing results 1 — 15 out of 97 results