Filters








2,700 Hits in 2.9 sec

Deep Tensor Encoding [article]

B Sengupta and E Vasquez and Y Qian
2017 arXiv   pre-print
be at par, in terms of average precision, with Fisher vector encoded image signatures.  ...  We illustrate a variety of feature encodings incl. sparse dictionary coding and Fisher vectors along with proposing that a structured tensor factorization scheme enables us to perform retrieval that can  ...  layer of a deep neural network.  ... 
arXiv:1703.06324v2 fatcat:ugf6ejv3n5cjtm23pqseb2mevy

Neural Codes for Image Retrieval [chapter]

Artem Babenko, Anton Slesarev, Alexandr Chigorin, Victor Lempitsky
2014 Lecture Notes in Computer Science  
In the experiments with several standard retrieval benchmarks, we establish that neural codes perform competitively even when the convolutional neural network has been trained for an unrelated classification  ...  Overall, our quantitative experiments demonstrate the promise of neural codes as visual descriptors for image retrieval.  ...  Fisher vectors).  ... 
doi:10.1007/978-3-319-10590-1_38 fatcat:s3ad4dk34jgdbawkkorr7wseba

Exploring Fisher vector and deep networks for action spotting

Zhe Wang, Limin Wang, Wenbin Du, Yu Qiao
2015 2015 IEEE Conference on Computer Vision and Pattern Recognition Workshops (CVPRW)  
Moreover, we exploit deep neural networks to extract both appearance and motion representation for this task.  ...  However, our current deep network fails to yield better performance than our Fisher vector based approach and may need further exploration.  ...  In the future, we may tune the convolutional neural networks and combine the with the Fisher vector representation to further improve the performance.  ... 
doi:10.1109/cvprw.2015.7301330 dblp:conf/cvpr/WangWD015 fatcat:5c7akoaambajfppowkaz7viaaa

Neural Codes for Image Retrieval [article]

Artem Babenko, Anton Slesarev, Alexandr Chigorin, Victor Lempitsky
2014 arXiv   pre-print
In the experiments with several standard retrieval benchmarks, we establish that neural codes perform competitively even when the convolutional neural network has been trained for an unrelated classification  ...  Overall, our quantitative experiments demonstrate the promise of neural codes as visual descriptors for image retrieval.  ...  Fisher vectors).  ... 
arXiv:1404.1777v2 fatcat:rzhl4fk5pbfljjlslxvkrbfn4q

Siamese Network of Deep Fisher-Vector Descriptors for Image Retrieval [article]

Eng-Jon Ong and Sameed Husain and Miroslaw Bober
2017 arXiv   pre-print
The CNN component produces dense, deep convolutional descriptors that are then aggregated by the Fisher Vector method.  ...  Crucially, we propose to simultaneously learn both the CNN filter weights and Fisher Vector model parameters.  ...  We achieve this using a novel combined CNN and Fisher Vector model that is learnt simultaneously.  ... 
arXiv:1702.00338v1 fatcat:yq4ki6zxcjfzdizxsw2z7xouvy

Remote Sensing Image Retrieval Using Convolutional Neural Network Features and Weighted Distance

Chippy Babu
2021 International Journal for Research in Applied Science and Engineering Technology  
A retrieval method supported weighted distance and basic features of Convolutional Neural Network (CNN) is proposed during this letter. the strategy contains two stages.  ...  also the retrieved images.  ...  Here we retrieve images using Convolutional Neural Network features and by calculating Weighted Distance.  ... 
doi:10.22214/ijraset.2021.37321 fatcat:mfoazrx3p5byvervsksuz7ybhi

Multilayer Convolutional Feature Aggregation Algorithm for Image Retrieval

Rongsheng Dong, Ming Liu, Fengying Li
2019 Mathematical Problems in Engineering  
The visual feature information in the convolutional neural network (CNN) is extracted, and the target response weight map is generated by combining with the spatial weighting algorithm of the target fuzzy  ...  First, the ML-RCroW algorithm inputs an image into the VGG16 (a deep convolutional neural network developed by researchers at Visual Geometry Group and Google DeepMind) network model in which the fully  ...  [19] proposed a method for extracting convolutional feature maps from different network layers and for using VLAD to encode features for image retrieval.  ... 
doi:10.1155/2019/9794202 fatcat:fi746ksiuzhlpjefeufybymhdu

Fisher Vectors Derived from Hybrid Gaussian-Laplacian Mixture Models for Image Annotation [article]

Benjamin Klein, Guy Lev, Gil Sadeh, Lior Wolf
2015 arXiv   pre-print
Finally, by using the new Fisher Vectors derived from HGLMMs, we achieve state-of-the-art results for both the image annotation and the image search by a sentence tasks.  ...  Motivated by the assumption that different distributions should be applied for different datasets, we present two other Mixture Models and derive their Expectation-Maximization and Fisher Vector expressions  ...  Acknowledgments This research is supported by the Intel Collaborative Research Institute for Computational Intelligence (ICRI-CI).  ... 
arXiv:1411.7399v2 fatcat:kjzlvvzcvfhjnirvpehdes6hx4

Object Classification using Ensemble of Local and Deep Features [article]

Siddharth Srivastava, Prerana Mukherjee, Brejesh Lall, Kamlesh Jaiswal
2017 arXiv   pre-print
We also compare and contrast effectiveness of feature representation capability of various layers of convolutional neural network.  ...  In this paper we propose an ensemble of local and deep features for object classification.  ...  Most of the techniques utilizing multiple convolutional neural networks require retraining the network.  ... 
arXiv:1712.04926v1 fatcat:vdcgkoexrrbd5atmglwevzvdtm

Modeling Text with Graph Convolutional Network for Cross-Modal Information Retrieval [article]

Jing Yu, Yuhang Lu, Zengchang Qin, Yanbing Liu, Jianlong Tan, Li Guo, Weifeng Zhang
2018 arXiv   pre-print
For cross-modal information retrieval between images and texts, existing work mostly uses off-the-shelf Convolutional Neural Network (CNN) for image feature extraction.  ...  A dual-path neural network model is proposed for couple feature learning in cross-modal information retrieval.  ...  Figure 2 : 2 The structure of the proposed model is a dual-path neural network: i.e., text Graph Convolutional Network (text GCN) (top) and image Neural Network (image NN) (bottom).  ... 
arXiv:1802.00985v2 fatcat:tcobdmh5qfcezmcj7oaevaz4gi

Fast content-based image retrieval using convolutional neural network and hash function

Domonkos Varga, Tamas Sziranyi
2016 2016 IEEE International Conference on Systems, Man, and Cybernetics (SMC)  
The success of deep learning techniques such as convolutional neural networks have motivated us to explore its applications in our context.  ...  Due to the explosive increase of online images, content-based image retrieval has gained a lot of attention.  ...  We are very thankful to Levente Kovács for helping us with professional advices in high-performance computing.  ... 
doi:10.1109/smc.2016.7844637 dblp:conf/smc/VargaS16 fatcat:vj5wgjkarzcxddfypouse3s6i4

Exploiting Local Features from Deep Networks for Image Retrieval [article]

Joe Yue-Hei Ng, Fan Yang, Larry S. Davis
2015 arXiv   pre-print
We show that for instance-level image retrieval, lower layers often perform better than the last layers in convolutional neural networks.  ...  We present an approach for extracting convolutional features from different layers of the networks, and adopt VLAD encoding to encode features into a single vector for each image.  ...  Other global features such as GIST descriptors and Fisher Vector (FV) [21] have also been evaluated for large-scale image retrieval.  ... 
arXiv:1504.05133v2 fatcat:jvivxjh2c5amxkoxl3p52riqfm

Exploiting local features from deep networks for image retrieval

Joe Yue-Hei Ng, Fan Yang, Larry S. Davis
2015 2015 IEEE Conference on Computer Vision and Pattern Recognition Workshops (CVPRW)  
We show that for instance-level image retrieval, lower layers often perform better than the last layers in convolutional neural networks.  ...  We present an approach for extracting convolutional features from different layers of the networks, and adopt VLAD encoding to encode features into a single vector for each image.  ...  Other global features such as GIST descriptors and Fisher Vector (FV) [21] have also been evaluated for large-scale image retrieval.  ... 
doi:10.1109/cvprw.2015.7301272 dblp:conf/cvpr/NgYD15 fatcat:kwgnyb2irfgtxhko6p3vg7dzze

What is the right way to represent document images? [article]

Gabriela Csurka, Diane Larlus, Albert Gordo, Jon Almazan
2016 arXiv   pre-print
descriptors, (2) deep features based on Convolutional Neural Networks, and (3) features extracted from hybrid architectures that take inspiration from the two previous ones.  ...  We propose a comprehensive experimental study that compares three types of visual document image representations: (1) traditional so-called shallow features, such as the RunLength and the Fisher-Vector  ...  Convolutional Neural Networks Convolutional Neural Networks (CNNs) are composed of several layers that combine linear as well as non-linear operators jointly learned, in an end-to-end manner, to solve  ... 
arXiv:1603.01076v3 fatcat:ugjusn6n6vhljn5j52uj5llzta

A Convolutional Neural Network-based Patent Image Retrieval Method for Design Ideation [article]

Shuo Jiang, Jianxi Luo, Guillermo Ruiz Pava, Jie Hu, Christopher L. Magee
2020 arXiv   pre-print
In turn, the trained neural network provides the deep features in the image embedding vectors that can be utilized for patent image retrieval and visual mapping.  ...  However, most patent mining research only focuses on textual information and ignores visual information. Herein, we propose a convolutional neural network (CNN)-based patent image retrieval method.  ...  ACKNOWLEDGEMENTS The authors acknowledge the funding support for this work received from the SUTD-MIT International Design Center and the China Scholarship Council (CSC).  ... 
arXiv:2003.08741v3 fatcat:jzjo2p375nb6helrhotdvfrpoq
« Previous Showing results 1 — 15 out of 2,700 results