Isotropic Hashing

By Weihao Kong, Wu-Jun Li. Neural Information Processing Systems, 2012

331 citations recorded by the source corpus.

View Paper: https://www.semanticscholar.org/paper/740be3f7d61c83068dbb85cf6bd209c8f550e6ea

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

Citations in the corpus

Discrete Graph Hashing (2014, methodology): “These include the unsupervised methods, e.g., Iterative Quantization [9], Isotropic Hashing [14], Spectral Hashing [38] [37], and Anchor Graph Hashing [23], the semi-supervised methods, e.g., Weakly-Supervised Hashing [25], and the supervised methods, e.g., Semantic Hashing [30], Binary Reconstructi”
Inter-media hashing for large-scale retrieval from heterogeneous data sources (2013, background): “Iterative quantization hashing [11] constructs hash functions by rotating the axes so that the difference between the binary codes and the projected data is minimized [11], while the goal of isotropic hashing [18] is to make the variances along the rotated linear projections are as same as possible.”
Learning Binary Codes for Maximum Inner Product Search (2015, methodology): “A number of algorithms in this category have been proposed, including unsupervised PCA Hashing [33], Iterative Quantization (ITQ) [8], Isotropic Hashing (IsoHash) [12], etc., and supervised Minimal Loss Hashing (MLH) [25, 26], Semi-Supervised Hashing (SSH) [33], Ranking-Based Su- 1550-5499/15 $31.00”
Supervised Deep Features for Software Functional Clone Detection by Exploiting Lexical and Syntactical Information in Source Code (2017, methodology): “Besides, we use learning to hash [Kong and Li, 2012] to further transform the real-valued representations to binary hash codes in the purpose of improving the detection efficiency and saving storage space.”
Learning Visual Semantic Relationships for Efficient Visual Retrieval (2015, methodology): “The representative linear algorithms contains unsupervised PCA Hashing [20], Iterative Quantization (ITQ) [19], Isotropic Hashing [29], etc.”
Fast Structural Binary Coding (2016, methodology): “…including one randomized method, Locality Sensitive Hashing (LSH) [Andoni and Indyk, 2008], one spectral approach, Spectral Hashing (SH) [Weiss et al., 2008], and two linear projection techniques, Iterative Quantization (ITQ) [Gong et al., 2012] and Isotropic Hashing (ISOH) [Kong and Li, 2012].”
Deep Cross-Modal Hashing (2016, methodology): “Furthermore, we can achieve a constant or sub-linear time complexity for search by using hash codes to construct an index [15].”
Progressive Generative Hashing for Image Retrieval (2018, methodology): “…learnt the projection based hash functions by leveraging the information contained in the data [Weiss et al. , 2008; Gong and Lazebnik, 2011; Wang et al. , 2017; Liu et al. , 2014b; Wang et al. , 2018; Kong and Li, 2012; Yu et al. , 2014; Liu et al. , 2014a; Shen et al. , 2015; Wang et al. , 2014].”
Feature Learning Based Deep Supervised Hashing with Pairwise Labels (2015, methodology): “…has become one of the most popular and effective techniques due to its fast query speed and low memory cost [Kulis and Grauman, 2009; Gong and Lazebnik, 2011; Kong and Li, 2012; Liu et al., 2012; Rastegari et al., 2013; He et al., 2013; Lin et al., 2014; Shen et al., 2015; Kang et al., 2016].”
Bit-Scalable Deep Hashing With Regularized Similarity Learning for Image Retrieval and Person Re-Identification (2015, background): “, the former bits are learned typically more significant than latter, so that one can simply pick desired number of bits fro m the top of the hashing codes [13][14][6][15].”
Supervised Discrete Hashing (2015, methodology): “The representative algorithms in this category include unsupervised PCA Hashing [34], Iterative Quantization (ITQ) [10], Isotropic Hashing [12], etc., and supervised Minimal Loss Hashing (MLH) [24, 25], Semi-Supervised Hashing (SSH) [34], LDA Hashing [2], Ranking-Based Supervised Hashing [35], FastH”
Learning to Hash for Indexing Big Data—A Survey (2015, background): “Other interesting learning ideas include twostep learning methods which treat hash bit learning and hash function learning separately [85][86].”
Sparse composite quantization (2015, methodology): “Hypercubic quantization, such as iterative quantization [7], isotropic hashing [16], harmonious hashing [32], angular quantization [6], can be regarded as a variant of scalar quantization by optimally rotating the data space and performing binary quantization along each dimension in the rotated spac”
Optimized Cartesian K-Means (2014, methodology): “According to the form of the hash function, we roughly categorize the binary encoding approaches as those based on Hamming embedding and on quantization.”
A Survey on Learning to Hash (2016, background): “Isotropic hashing [63] finds a rotation following PCA preprocessing such that R>VV>R 1⁄4 S becomes a matrix with equal diagonal values, i.”
Hashing for Similarity Search: A Survey (2014, background): “In the case that the true data distribution is a multi-dimensional isotropic Gaussian distribution, it is equivalent to iterative quantization [30], [31] and isotropic hashing [60].”
Optimized Product Quantization (2014, methodology): “A recent method called isotropic hashing [28] explicitly optimizes the balance.”
Top Rank Supervised Binary Coding for Visual Search (2015, methodology): “For those unsupervised algorithms, we observe that SH, ITQ and ISOH consistently and significantly outperform LSH.”
Online sketching hashing (2015, methodology): “Representative works include Iterative Quantization (ITQ) [9], Isotropic Hashing (IsoH) [17], Spherical Hashing (SpH) [12] and so on [25, 19, 24].”