Compact Global Descriptors for Visual Search

Chandrasekhar, Vijay; Lin, Jie; Morère, Olivier; Veillard, Antoine; Goh, Hanlin

doi:10.1109/dcc.2015.54

Cited by 12 publications

(18 citation statements)

References 24 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…In this experiment (Table V) we compare the behavior of retrieval performance for different lengths of the hash code (for m-kmeans-t 1 ) and for different values of n nearest neighbors (for [60] 0.381 0.225 -PCAHash [54] 0.528 0.239 -LSH [61] 0.431 0.239 -SKLSH [62] 0.241 0.134 -SH [2] 0.522 0.232 -SRBM [63] 0.516 0.212 -UTH [31] 0.571 0.240 -m-k-means-n 1 (n = m-kmeans-n 1 ). Experiments were made for SIFT1M for different values of recall@R. Fig.…”

Section: Results Varying Hash Code Length and Nmentioning

confidence: 99%

Compact Hash Codes for Efficient Visual Descriptors Retrieval in Large Scale Databases

Ercoli

Bertini

Bimbo

2017

IEEE Trans. Multimedia

View full text Add to dashboard Cite

Abstract-In this paper, we present an efficient method for visual descriptors retrieval based on compact hash codes computed using a multiple k-means assignment. The method has been applied to the problem of approximate nearest neighbor (ANN) search of local and global visual content descriptors, and it has been tested on different datasets: three large scale standard datasets of engineered features of up to one billion descriptors (BIGANN) and, supported by recent progress in convolutional neural networks (CNNs), on CIFAR-10, MNIST, INRIA Holidays, Oxford 5K, and Paris 6K datasets; also, the recent DEEP1B dataset, composed by one billion CNN-based features, has been used. Experimental results show that, despite its simplicity, the proposed method obtains a very high performance that makes it superior to more complex state-ofthe-art methods.Index Terms-Convolutional neural network (CNN), hashing, nearest neighbor search, retrieval, SIFT.

show abstract

Section: Results Varying Hash Code Length and Nmentioning

confidence: 99%

Compact Hash Codes for Efficient Visual Descriptors Retrieval in Large Scale Databases

Ercoli

Bertini

Bimbo

2017

IEEE Trans. Multimedia

View full text Add to dashboard Cite

show abstract

“…To evaluate the effect of network weights initialization in the pre-training stage, we present mAP results on Holidays dataset at output size 64 bits (see Figure 2(b)), for comparison (1) SRBM based network weights proposed in our previous work [15] with (2) random unit-norm network weights (denoted as UniW). In addition, we report the results combining the proposed UTH scheme with either SRBM (denoted as UTH SRBM) or UniW (denoted as UTH UniW).…”

Section: Resultsmentioning

confidence: 99%

“…Here, we progressively decrease the dimensionality of hidden layers by a factor of 2, and train several RBMs with varying number of hidden layers and output units to optimize parameters. More details are available in our previous work [15].…”

Section: Unsupervised Triplet Hashingmentioning

confidence: 99%

See 1 more Smart Citation

Tiny Descriptors for Image Retrieval with Unsupervised Triplet Hashing

Lin

Morère

Petta³

et al. 2016

2016 Data Compression Conference (DCC)

Self Cite

View full text Add to dashboard Cite

A typical image retrieval pipeline starts with the comparison of global descriptors from a large database to find a short list of candidate matches. A good image descriptor is key to the retrieval pipeline and should reconcile two contradictory requirements: providing recall rates as high as possible and being as compact as possible for fast matching. Following the recent successes of Deep Convolutional Neural Networks (DCNN) for large scale image classification, descriptors extracted from DCNNs are increasingly used in place of the traditional hand crafted descriptors such as Fisher Vectors (FV) with better retrieval performances. Nevertheless, the dimensionality of a typical DCNN descriptor -extracted either from the visual feature pyramid or the fully-connected layers-remains quite high at several thousands of scalar values.In this paper, we propose Unsupervised Triplet Hashing (UTH), a fully unsupervised method to compute extremely compact binary hashes -in the 32-256 bits range-from high-dimensional global descriptors. UTH consists of two successive deep learning steps. First, Stacked Restricted Boltzmann Machines (SRBM), a type of unsupervised deep neural nets, are used to learn binary embedding functions able to bring the descriptor size down to the desired bitrate. SRBMs are typically able to ensure a very high compression rate at the expense of loosing some desirable metric properties of the original DCNN descriptor space. Then, triplet networks, a rank learning scheme based on weight sharing nets is used to fine-tune the binary embedding functions to retain as much as possible of the useful metric properties of the original space. A thorough empirical evaluation conducted on multiple publicly available dataset using DCNN descriptors shows that our method is able to significantly outperform state-of-the-art unsupervised schemes in the target bit range. * Equal contributions from Jie Lin, Olivier Morère and Julie Petta.

show abstract

“…Holidays Oxford 5K Paris 6K ITQ [11] 53.68 23.00 -BPBC [10] 38.10 22.51 -PCAHash [11] 52.80 23.90 -LSH [6] 43.08 23.91 -SKLSH [26] 24.09 13.39 -SH [32] 52. 22 23.24 -SRBM [4] 51.58 21.23 -UTH [20] 57. 10 In the second experiment we evaluate a use case in which a database of images is queried with a large number of images that do not belong to it.…”

Section: Methodsmentioning

confidence: 99%

Bloom Filters and Compact Hash Codes for Efficient and Distributed Image Retrieval

Salvi¹,

Ercoli

Bertini

et al. 2016

2016 IEEE International Symposium on Multimedia (ISM)

View full text Add to dashboard Cite

Abstract-This paper presents a novel method for efficient image retrieval, based on a simple and effective hashing of CNN features and the use of an indexing structure based on Bloom filters. These filters are used as gatekeepers for the database of image features, allowing to avoid to perform a query if the query features are not stored in the database and speeding up the query process, without affecting retrieval performance. Thanks to the limited memory requirements the system is suitable for mobile applications and distributed databases, associating each filter to a distributed portion of the database (database shard), addressing large scale archives and allowing query parallelization. Experimental validation has been performed on three standard image retrieval datasets, outperforming state-of-the-art hashing methods in terms of precision, while the proposed indexing method obtains a 2× speedup.

show abstract

Compact Global Descriptors for Visual Search

Cited by 12 publications

References 24 publications

Compact Hash Codes for Efficient Visual Descriptors Retrieval in Large Scale Databases

Compact Hash Codes for Efficient Visual Descriptors Retrieval in Large Scale Databases

Tiny Descriptors for Image Retrieval with Unsupervised Triplet Hashing

Bloom Filters and Compact Hash Codes for Efficient and Distributed Image Retrieval

Contact Info

Product

Resources

About