Yannis Avrithis, Giorgos Tolias, Yannis Kalantidis · DSpace - NTUA (National Technical University of Athens) · 2010
We present a new approach to image indexing and retrieval, which integrates appearance with global image geometry in the indexing process, while enjoying robustness against viewpoint change, photometric variations, occlusion, and background clutter. We exploit shape parameters of local features to estimate image alignment via a single correspondence. Then, for each feature, we construct a sparse spatial map of all remaining features, encoding their normalized position and appearance, typically vector quantized to visual word. An image is represented by a collection of such feature maps and RANSAC-like matching is reduced to a number of set intersections. Because the induced dissimilarity is still not a metric, we extend min-wise independent permutations to collections of sets and derive a similarity measure for feature map collections. We then exploit sparseness to build an inverted file whereby the retrieval process is sub-linear in the total number of images, ideally linear in the number of relevant ones. We achieve excellent performance on 10^4 images, with a query time in the order of milliseconds.
Cited by 22Open access
Tuan Hoang, Thanh-Toan Do, Dang-Khoa Le Tan et al. · arXiv (Cornell University) · 2017
We introduce a novel approach to improve unsupervised hashing. Specifically, we propose a very efficient embedding method: Gaussian Mixture Model embedding (Gemb). The proposed method, using Gaussian Mixture Model, embeds feature vector into a low-dimensional vector and, simultaneously, enhances the discriminative property of features before passing them into hashing. Our experiment shows that the proposed method boosts the hashing performance of many state-of-the-art, e.g. Binary Autoencoder (BA) [1], Iterative Quantization (ITQ) [2], in standard evaluation metrics for the three main benchmark datasets.
Cited by 0Open access
Qixiang Wang, Shanfeng Wang, Maoguo Gong et al. · 2018
The goal of network representation learning is to embed nodes so as to encode the proximity structures of a graph into a continuous low-dimensional feature space. In this paper, we propose a novel algorithm called node2hash based on feature hashing for generating node embeddings. This approach follows the encoder-decoder framework. There are two main mapping functions in this framework. The first is an encoder to map each node into high-dimensional vectors. The second is a decoder to hash these vectors into a lower dimensional feature space. More specifically, we firstly derive a proximity measurement called expected distance as target which combines position distribution and co-occurrence statistics of nodes over random walks so as to build a proximity matrix, then introduce a set of T different hash functions into feature hashing to generate uniformly distributed vector representations of nodes from the proximity matrix. Compared with the existing state-of-the-art network representation learning approaches, node2hash shows a competitive performance on multi-class node classification and link prediction tasks on three real-world networks from various domains.
Cited by 11Open access
Cornelia Caragea, Adrian Silvescu, Prasenjit Mitra · 2011
Recent advances in next-generation sequencing technologies have resulted in an exponential increase in protein sequence data. The k-gram representation, used for protein sequence classification, usually results in prohibitively high dimensional input spaces, for large values of k. Applying data mining algorithms to these input spaces may be intractable due to the large number of dimensions. Hence, using dimensionality reduction techniques can be crucial for the performance and the complexity of the learning algorithms. We study the applicability of feature hashing to protein sequence classification, where the original high-dimensional space is "reduced" by mapping features to hash keys, such that multiple features can be mapped (at random) to the same key, and "aggregating" their counts. We compare feature hashing with the "bag of k-grams" and feature selection approaches. Our results show that feature hashing is an effective approach to reducing dimensionality on protein sequence classification tasks.
Cited by 22
Cornelia Caragea, Adrian Silvescu, Prasenjit Mitra · Proteome Science · 2012
Recent advances in next-generation sequencing technologies have resulted in an exponential increase in the rate at which protein sequence data are being acquired. The k-gram feature representation, commonly used for protein sequence classification, usually results in prohibitively high dimensional input spaces, for large values of k. Applying data mining algorithms to these input spaces may be intractable due to the large number of dimensions. Hence, using dimensionality reduction techniques can be crucial for the performance and the complexity of the learning algorithms. In this paper, we study the applicability of feature hashing to protein sequence classification, where the original high-dimensional space is "reduced" by hashing the features into a low-dimensional space, using a hash function, i.e., by mapping features into hash keys, where multiple features can be mapped (at random) to the same hash key, and "aggregating" their counts. We compare feature hashing with the "bag of k-grams" approach. Our results show that feature hashing is an effective approach to reducing dimensionality on protein sequence classification tasks.
Cited by 49Open access
Dan Shi, Lei Zhu, Jingjing Li et al. · IEEE Transactions on Image Processing · 2023
Unsupervised feature selection chooses a subset of discriminative features to reduce feature dimension under the unsupervised learning paradigm. Although lots of efforts have been made so far, existing solutions perform feature selection either without any label guidance or with only single pseudo label guidance. They may cause significant information loss and lead to semantic shortage of the selected features as many real-world data, such as images and videos are generally annotated with multiple labels. In this paper, we propose a new Unsupervised Adaptive Feature Selection with Binary Hashing (UAFS-BH) model, which learns binary hash codes as weakly-supervised multi-labels and simultaneously exploits the learned labels to guide feature selection. Specifically, in order to exploit the discriminative information under the unsupervised scenarios, the weakly-supervised multi-labels are learned automatically by specially imposing binary hash constraints on the spectral embedding process to guide the ultimate feature selection. The number of weakly-supervised multi-labels (the number of "1" in binary hash codes) is adaptively determined according to the specific data content. Further, to enhance the discriminative capability of binary labels, we model the intrinsic data structure by adaptively constructing the dynamic similarity graph. Finally, we extend UAFS-BH to multi-view setting as Multi-view Feature Selection with Binary Hashing (MVFS-BH) to handle the multi-view feature selection problem. An effective binary optimization method based on the Augmented Lagrangian Multiple (ALM) is derived to iteratively solve the formulated problem. Extensive experiments on widely tested benchmarks demonstrate the state-of-the-art performance of the proposed method on both single-view and multi-view feature selection tasks. For the purpose of reproducibility, we provide the source codes and testing datasets at https://github.com/shidan0122/UMFS.git..
Cited by 60
Kilian Q. Weinberger, Anirban Dasgupta, Josh Attenberg et al. · arXiv (Cornell University) · 2009
Empirical evidence suggests that hashing is an effective strategy for dimensionality reduction and practical nonparametric estimation. In this paper we provide exponential tail bounds for feature hashing and show that the interaction between random subspaces is negligible with high probability. We demonstrate the feasibility of this approach with experimental results for a new use case -- multitask learning with hundreds of thousands of tasks.
Cited by 16Open access
Bin Zhao, Eric P. Xing · Figshare · 2014
Curse of dimensionality is a practical and challenging problem in image categorization, especially in cases with a large number of classes. Multi-class classification encounters severe computational and storage problems when dealing with these large scale tasks. In this paper, we propose hierarchical feature hashing to effectively reduce dimensionality of parameter space without sacrificing classification accuracy, and at the same time exploit information in semantic taxonomy among categories. We provide detailed theoretical analysis on our proposed hashing method. Moreover, experimental results on object recognition and scene classification further demonstrate the effectiveness of hierarchical feature hashing.
Cited by 9Open access
Zhao, Bin, P Xing, Eric · KiltHub Repository · 2018
Curse of dimensionality is a practical and challenging problem in image categorization, especially in cases with a large number of classes. Multi-class classification encounters severe computational and storage problems when dealing with these large scale tasks. In this paper, we propose hierarchical feature hashing to effectively reduce dimensionality of parameter space without sacrificing classification accuracy, and at the same time exploit information in semantic taxonomy among categories. We provide detailed theoretical analysis on our proposed hashing method. Moreover, experimental results on object recognition and scene classification further demonstrate the effectiveness of hierarchical feature hashing.
Cited by 0Open access
Wei Wu, Bin Li, Ling Chen et al. · Lecture notes in computer science · 2016
Cited by 2
Kilian Q. Weinberger, Anirban Dasgupta, John Langford et al. · 2009
Empirical evidence suggests that hashing is an effective strategy for dimensionality reduction and practical nonparametric estimation. In this paper we provide exponential tail bounds for feature hashing and show that the interaction between random subspaces is negligible with high probability. We demonstrate the feasibility of this approach with experimental results for a new use case --- multitask learning with hundreds of thousands of tasks.
Cited by 928
Tongtong Yuan, Weihong Deng, Jiani Hu et al. · Neurocomputing · 2018
Cited by 19
Kang Chen, Li Zhu, Xueming Qian · Lecture notes in computer science · 2016
Cited by 1
Zifeng Huang, Yuxing Li, Qikang Zhang · Applied Soft Computing · 2025
Cited by 0
Xiao-Jiao Mao, Zhenfei Ju, Rui Xu et al. · 2015
Binary code learning techniques have been actively studied for hashing based nearest neighbor search in recent years. However, most existing techniques directly map the data into a Hamming space, which ignores the inherent property that original features may lie in different subspaces. To address this issue, this paper proposes a novel method by learning binary codes on the latent components decomposed from the original features of the data. We assume that each latent component represents an underlying feature subspace and similar/dissimilar data may contain the same/different latent components. As a result, the decomposition step distinguishes the original data better and consequently improves the discriminative power of binary codes. The experimental results of image retrieval tasks carried out on the commonly used benchmark datasets show that the proposed method outperforms other state-of-the-art methods on generating similarity-preserving binary codes.
Cited by 0
Artem Sokolov, Stefan Riezler · 2015
Randomly hashing multiple features into one aggregated feature is routinely used in large-scale machine learning tasks to both increase speed and decrease memory requirements, with little or no sacrifice in performance. In this paper we investigate whether using a learned (instead of a random) hashing function improves performance. We showed experimentally that with increasing difference between the dimensionalities of the input space and the hashed space, learning hashes is increasingly useful compared to random hashing. 1
Cited by 0
Rameshwar Pratap, Suryakant Bhardwaj, Hrushikesh Sudam Sarode et al. · 2021 IEEE International Conference on Big Data (Big Data) · 2021
Feature hashing algorithms [14], [21] are well-known algorithmic techniques for handling large dimensionality of the datasets. These methods reduce the dimensionality of data points while preserving the pairwise distance, or similarity between them. However, to the best of our knowledge, these feature hashing algorithms are not adaptable to the dynamic insertion and deletion of features. In this work, we suggest algorithms using which the existing feature hashing algorithms can be made adapted to dynamic feature insertion and deletion. Our algorithms fit in the framework of both real-valued [21], and binary [14] feature hashing algorithms. We show a theoretical analysis of our algorithms, and complement it with experiments on real-world datasets. Our algorithms suggest comparable accuracy w.r.t. baselines while simultaneously offering a significant speed-up in the running time. Our proposal is easy to implement and can be adopted in practice.
Cited by 0
Federico Tombari, Luigi Di Stefano · 2011
Geometric Hashing is a well-known technique for object recognition. This paper proposes a novel method aimed at improving the performance of Geometric Hashing in terms of robustness toward occlusion and clutter. To this purpose, it employs feature descriptors to notably decrease the amount of false positives that generally arise under these conditions. An additional advantage of the proposed technique with respect to the original method is the reduction of the computation requirements, which becomes significant with increasing number
Cited by 0Open access
Xiaolu Yu · 2024
Due to the rapid development of deep learning, the iris feature extraction performance has been significantly enhanced. However, iris features extracted by convolutional neural networks are typically represented as floating-point numbers. This results in degradation in data storage and retrieval performance. To this end, this paper proposes an iris feature extraction method. This method enhances feature discrimination to improve iris recognition capability and utilizes binary hash codes to enhance storage and retrieval efficiency. Firstly, category labels are mapped to hash centers. Subsequently, the feature vectors extracted by the convolutional neural network gradually approach the hash centers under the influence of the loss function, the feature vectors exhibit a clustered distribution in the hash space. Finally, the convolutional neural network converts iris images into binary hash codes for iris recognition. Experiments on two public iris datasets validate the effectiveness of the proposed method in iris verification and quick data storage and retrieval outcomes.
Cited by 0
Jin-Yu Liu, Xian-Ling Mao, Wenjin Ji et al. · IEEE Transactions on Circuits and Systems for Video Technology · 2026
Cited by 0