Filters








216,783 Hits in 6.2 sec

Features for image retrieval: an experimental comparison

Thomas Deselaers, Daniel Keysers, Hermann Ney
2007 Information retrieval (Boston)  
An experimental comparison of a large number of different image descriptors for content-based image retrieval is presented.  ...  In this paper , we first give an overview of a large variety of features for content-based image retrieval and compare them quantitatively on four different tasks: stock photo retrieval, personal photo  ...  The authors would like to thank Gyuri Dorkó (formerly with IN-RIA Rhône-Alpes) for providing his SIFT feature extraction software and the authors of the MPEG7 XM reference implementation.  ... 
doi:10.1007/s10791-007-9039-3 fatcat:sttrquiwvrcztm5kqz3ay73xdm

Moment Features Weighting for Image Retrieval

HabibollahAgh Atabay
2016 IOSR Journal of Computer Engineering  
Feature selection is an effective tool to improve the performance of content based image retrieval systems.  ...  This paper presents an effective moment weighting method according to image reconstruction and retrieval accuracyto reduce the dimensionality of moment-based features.  ...  Also this selection improves the retrieval accuracy, in comparison with the retrieval without feature selection.  ... 
doi:10.9790/0661-1805043237 fatcat:z3bprmstyfewnmd3v3qr476b24

Local Image Features for Shoeprint Image Retrieval

H. Su, D. Crookes, A. Bouridane, M. Gueham
2007 Procedings of the British Machine Vision Conference 2007  
(ii). for each feature detected, an enhanced SIFT descriptor is computed to represent this feature.  ...  This paper deals with the retrieval of scene-of-crime (or scene) shoeprint images from a reference database of shoeprint images by using a new local feature detector and an improved local feature descriptor  ...  Then, for each local feature, an enhanced SIFT [18] descriptor is computed to represent the feature.  ... 
doi:10.5244/c.21.38 dblp:conf/bmvc/SuCBG07 fatcat:rl5jng5l4fagpld5svqcjpcxeq

Independent Feature Analysis for Image Retrieval [chapter]

Jing Peng, Bir Bhanu
1999 Lecture Notes in Computer Science  
Content-based image retrieval methods based on the Euclidean metric expect the feature space to be isotropic.  ...  They suer from unequal dierential relevance of features in computing the similarity between images in the input feature space.  ...  Simple K nearest neighbor search, as an image retrieval procedure, returns the K images closest to the query.  ... 
doi:10.1007/3-540-48097-8_9 fatcat:6r53iu4t3ve2fkhzu2gxfclj6a

Independent feature analysis for image retrieval

Jing Peng, Bir Bhanu
2001 Pattern Recognition Letters  
Content-based image retrieval methods based on the Euclidean metric expect the feature space to be isotropic.  ...  They suer from unequal dierential relevance of features in computing the similarity between images in the input feature space.  ...  Simple K nearest neighbor search, as an image retrieval procedure, returns the K images closest to the query.  ... 
doi:10.1016/s0167-8655(00)00100-8 fatcat:q3rklmbvxzaujpwn25t5eddhuy

Learning Super-Features for Image Retrieval [article]

Philippe Weinzaepfel, Thomas Lucas, Diane Larlus, Yannis Kalantidis
2022 arXiv   pre-print
In this paper, we propose a novel architecture for deep image retrieval, based solely on mid-level features that we call Super-features.  ...  For training, they require only image labels. A contrastive loss operates directly at the level of Super-features and focuses on those that match across images.  ...  Overall, for either day or night images, FIRe either improves or performs on par to both methods compared.  ... 
arXiv:2201.13182v1 fatcat:qbovzqagkrbihkfeszxxilohkm

Deep Feature Aggregation and Image Re-ranking with Heat Diffusion for Image Retrieval [article]

Shanmin Pang and Jin Ma and Jianru Xue and Jihua Zhu and Vicente Ordonez
2018 arXiv   pre-print
Image retrieval based on deep convolutional features has demonstrated state-of-the-art performance in popular benchmarks.  ...  A distinctive problem in image retrieval is that repetitive or bursty features tend to dominate final image representations, resulting in representations less distinguishable.  ...  [34] introduced an unsupervised fine-tuning of CNN for image retrieval from a large collection of unordered images in a fully automated manner.  ... 
arXiv:1805.08587v5 fatcat:4xfc62r7e5cglfqhq6j5l3aoii

Selective Deep Convolutional Features for Image Retrieval [article]

Tuan Hoang, Thanh-Toan Do, Dang-Khoa Le Tan, Ngai-Man Cheung
2017 arXiv   pre-print
Convolutional Neural Network (CNN) is a very powerful approach to extract discriminative local descriptors for effective image search.  ...  Firstly, we propose various masking schemes, namely SIFT-mask, SUM-mask, and MAX-mask, to select a representative subset of local convolutional features and remove a large number of redundant features.  ...  Comparison to the state of the art We thoroughly compare our proposed framework with state-of-art methods in image retrieval task. We report experimental results in Table 6 .  ... 
arXiv:1707.00809v2 fatcat:xur4x7ggwrbizf2qkgys66pove

Aggregating Deep Convolutional Features for Image Retrieval [article]

Artem Babenko, Victor Lempitsky
2015 arXiv   pre-print
In this paper we investigate possible ways to aggregate local deep features to produce compact global descriptors for image retrieval.  ...  Several recent works have shown that image descriptors produced by deep convolutional neural networks provide state-of-the-art performance for image classification and retrieval problems.  ...  based on deep features and, in fact, over previous state-of-the-art for compact global image descriptors.  ... 
arXiv:1510.07493v1 fatcat:e6khopiv2rclhnzsjg4gliheyi

Features for Image Retrieval: A Quantitative Comparison [chapter]

Thomas Deselaers, Daniel Keysers, Hermann Ney
2004 Lecture Notes in Computer Science  
We compare the features for two different image retrieval tasks (color photographs and medical radiographs) and a clear difference in performance is observed, which can be used as a basis for an appropriate  ...  In this paper, different well-known features for image retrieval are quantitatively compared and their correlation is analyzed.  ...  In this work, the goal is not to introduce new features but to give quantitative results for a comparison of existing features for image retrieval tasks.  ... 
doi:10.1007/978-3-540-28649-3_28 fatcat:gf5elat43vclpj7bfzsccervgq

PLT-based spectral features for texture image retrieval

Joydeb Kumar Sana, Md. Monirul Islam
2018 IET Image Processing  
Effective texture feature is an essential component in any content-based image retrieval system. In this study, new texture features based on image enhancement technique are presented.  ...  The experimental results confirm that the proposed features have more tolerance to scale, orientation and illumination distortion than the state-ofthe-art Gabor, curvelet, Gaussian copula models of Gabor  ...  curvelet features and recent curvelet features on an original Brodatz image databaseFig. 12 Comparison of retrieval performance on scale distorted image database (a) Comparison of retrieval performance  ... 
doi:10.1049/iet-ipr.2018.5604 fatcat:hld27tk2uvaclhqrmrgfwkx5o4

Color Features Performance Comparison for Image Retrieval [chapter]

Daniele Borghesani, Costantino Grana, Rita Cucchiara
2009 Lecture Notes in Computer Science  
This paper proposes a comparison of color features for image retrieval. In particular the UCID image database has been employed to compare the retrieval capabilities of different color descriptors.  ...  The set of descriptors comprises global and spatially related features, and the tests show that HSV based global features provide the best performance at varying brightness and contrast settings.  ...  Experimental Results The database used for our test was UCIDv.2 (UnCompressed Image Database), provided by [17] .  ... 
doi:10.1007/978-3-642-04146-4_96 fatcat:nkzmbljvxjgnvhtx7x2o7yzmcy

Aggregating Local Deep Features for Image Retrieval

Artem Babenko Yandex, Victor Lempitsky
2015 2015 IEEE International Conference on Computer Vision (ICCV)  
In this paper we investigate possible ways to aggregate local deep features to produce compact global descriptors for image retrieval.  ...  Several recent works have shown that image descriptors produced by deep convolutional neural networks provide state-of-the-art performance for image classification and retrieval problems.  ...  For most retrieval datasets, objects of interest tend to be located close to the geometrical center of an image.  ... 
doi:10.1109/iccv.2015.150 dblp:conf/iccv/BabenkoL15 fatcat:y6zohrloarfjroswjgcsxbugzm

Selective Deep Convolutional Features for Image Retrieval

Tuan Hoang, Thanh-Toan Do, Dang-Khoa Le Tan, Ngai-Man Cheung
2017 Proceedings of the 2017 ACM on Multimedia Conference - MM '17  
Convolutional Neural Network (CNN) is a very powerful approach to extract discriminative local descriptors for e ective image search.  ...  Firstly, we propose various masking schemes, namely SIFT-mask, SUM-mask, and MAX-mask, to select a representative subset of local convolutional features and remove a large number of redundant features.  ...  Comparison to the state of the art We thoroughly compare our proposed framework with state-of-art methods in image retrieval task. We report experimental results in Table 6 .  ... 
doi:10.1145/3123266.3123417 dblp:conf/mm/HoangDTC17 fatcat:gnevsc2oafa7pkdl3nkc4qenge

Deep Bottleneck Feature for Image Classification

Yan Song, Ian McLoughLin, Lirong Dai
2015 Proceedings of the 5th ACM on International Conference on Multimedia Retrieval - ICMR '15  
Effective image representation plays an important role for image classification and retrieval. Bag-of-Features (BoF) is well known as an effective and robust visual representation.  ...  In this paper, we propose a bag of Deep Bottleneck Features (DBF) for image classification, effectively combining the strengths of a CNN within a BoF framework.  ...  INTRODUCTION Effective image representation plays an important role in content based recognition and retrieval applications.  ... 
doi:10.1145/2671188.2749314 dblp:conf/mir/SongMD15 fatcat:s3nzs7srbfgbblizxge3xeuz5e
« Previous Showing results 1 — 15 out of 216,783 results