Spectral Hashing

By Yair Weiss, Antonio Torralba, Rob Fergus. Neural Information Processing Systems, 2008

619 citations recorded by the source corpus.

View Paper: https://www.semanticscholar.org/paper/8a4fb7c39fde7ae172f35cd68a802de168cd4466

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

Multiple feature hashing for real-time large scale near-duplicate video retrieval (2011, methodology): “Compared with the hashing algorithms proposed in [27, 35] where only local information is considered, it may eliminate the problem of overﬁtting, making the learning more robust.”
Fast GPU-based locality sensitive hashing for k-nearest neighbor computation (2011, unknown): “For example, the preﬁx tree used in LSH-forest [3] and the PCA used in spectral hashing [43] are not the best candidates for GPU parallelization.”
Two-Stream Deep Hashing With Class-Specific Centers for Supervised Image Search (2020, unknown): “3) SH [38] treats the problem of finding the best code for a given data set as graph partitioning problem and obtains a spectral method by relaxing this partitioning problem.”
A Survey of Multi-View Representation Learning (2016, unknown): “Kumar and Udupa [146] extend Spectral Hashing [158] from the single view setting to the multi-view scenario and present cross view hashing (CVH), which attempts to find hash functions that map similar objects to similar codewords over all the views so that inter-view and intraview similarities are b”
Semantics-preserving hashing for cross-view retrieval (2015, methodology): “Existing hashing methods can be roughly classiﬁed into two categories: single-view hashing [16, 21, 4, 11, 6, 13, 8, 22, 10] and multi-view hashing [1, 9, 20, 24, 23, 18, 27, 14, 3, 25, 17, 26, 19], with the former focusing on data with a single view while the latter focusing on that with multiple…”
Packing bag-of-features (2009, background): “[13] which reduces the global GIST descriptor to a small binary code with spectral”
Searching in one billion vectors: Re-rank with source coding (2011, unknown): “But as shown in [5], this post verification is also important for m ethods based on binary [4, 15] or quantized codes [11, 5, 6], as the ra nking provided on output of the large scale search is significantly improved when verifying the few first hypotheses.”
Optimized Product Quantization (2014, unknown): “This also happens in the derivations of spectral hashing (SH) [16].”
Joint-modal Distribution-based Similarity Hashing for Large-scale Unsupervised Deep Cross-modal Retrieval (2020, unknown): “As two of the earliest unsupervised cross-modal hashing methods, both Cross-view hashing (CVH) [17] and Inter-Media Hashing (IMH) [29] extend Spectral Hashing [37] from unimodal hashing to cross-modal hashing.”
Structure-Preserving Binary Representations for RGB-D Action Recognition (2016, methodology): “Another popular technique called Spectral Hashing (SpH) [57] was also proposed to preserve the locality information of data.”
Discrete Collaborative Filtering (2016, background): “Second, as similarity calculation by inner product in a vector space is replaced by bit operations in a proper Hamming space, the time complexity of linear scan is signiﬁcantly reduced and even constant time search is made possible by exploiting lookup tables [28, 31].”
Learning to Hash with Binary Reconstructive Embeddings (2009, methodology): “These include semantic hashing [3], spectral hashin g [4], parameter-sensitive hashing [13], and boosting-based hashing methods [14].”
Fast Approximate Nearest Neighbor Search With The Navigating Spreading-out Graph (2017, methodology): “For dense continuous vectors, various solutions have been proposed such as tree-structure based approaches [2,6,8,17,24,36], hashing-based approaches [18, 20, 23, 32, 40], quantization-based approaches [1, 19, 26, 39], and graph-based approaches [3, 21, 33, 41].”
Kernelized locality-sensitive hashing for scalable image search (2009, methodology): “Recent work has considered how to construct or learn an embedding that will preserve the desired distance function, typically with the intention of mapping to a very low-dimensional space that is more easily searchable with known techniques [1, 20, 28, 22, 31].”
Discriminative coupled dictionary hashing for fast cross-media retrieval (2014, methodology): “…two retrieval schemes, we compare with the state-of-the-art cross-media hashing methods CMSSH [3], CVH [9], MLBE [28], LCMH [30]; for the third retrieval scheme, we additionally compare with some uni-modal hashing approaches: SH [23], KLSH [8], AGH [11] and a multi-view hashing approach MFH [17].”
Negative evidences and co-occurrences in image retrieval: the benet of PCA and whitening (2012, methodology): “In this paper, we will mainly focus on more scalable approaches in the spirit of recent work on compact images representations [6–8], where the image description is a short vector, which is subsequently encoded in compact codes using binarization [6, 9] or product quantization techniques [10].”
Searching with quantization: approximate nearest neighbor search using short codes and distance estimators (2009, methodology): “This is possible in the context of a global descriptor [13] and [15].”
Person Re-identification: Past, Present and Future (2016, background): “Learning to hash is popular in the community following the milestone work Spectral Hashing [150].”
Locality-sensitive binary codes from shift-invariant kernels (2009, methodology): “To demonstrate the basic behavior of our method, we first report results for two-dimensional synthetic data using a protocol similar to [15] (we have also conducted tests on higher-dimensional synthetic data, with very similar results).”