A contextual dissimilarity measure for accurate and efficient image search

By Herv'e J'egou, Hedi Harzallah, Cordelia Schmid. 2007

298 citations recorded by the source corpus.

Subfields: Artificial Intelligence, Computer Vision and Pattern Recognition, Information Retrieval & Search

How researchers see this paper

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

In Artificial Intelligence

In Computer Vision and Pattern Recognition

In Information Retrieval & Search

Citations in the corpus

Product Quantization for Nearest Neighbor Search (2011, methodology): “To address this problem, we use the multiple assignment strategy of [27].”
Contextual Hausdorff dissimilarity for multi-instance clustering (2012, unknown): “Different from [3], the contextual information [4] used is not only the neighborhood of each point but also the global information of the data set.”
Hello neighbor: Accurate object retrieval with k-reciprocal nearest neighbors (2011, background): “However unlike [12, 13], we do not directly symmetrize nearest neighborhood relationships in this work.”
Lost in quantization: Improving particular object retrieval in large scale image databases (2008, methodology): “dissimilarity measure” [9] was found to improve retrieval performance on the image retrieval benchmark of [14], but requires computing k-nearest neighbouring images for each image in the database, which may become prohibitively expensive on very large image collections.”
Re-ranking Person Re-identification with k-Reciprocal Encoding (2017, methodology): “In literature, the �-reciprocal nearest neighbor [15, 35] is an effective solution to the above-mentioned problem, i.”
Near Duplicate Image Detection: min-Hash and tf-idf Weighting (2008, background): “Recently, most of the successful image indexing approaches are based on the bag-ofvisual-words representation [5, 9, 17, 18, 19].”
Total recall II: Query expansion revisited (2011, background): “Virtually all aspects of particular object BoW-type retrieval have been intensively studied: feature detectors and descriptors [14, 5, 25, 15, 16], vocabulary construction [23, 18, 21, 9, 17], spatial verification and re-ranking [21, 9], document metric learning [12, 10, 6] and dimensionality reduct”
A Pose-Sensitive Embedding for Person Re-identification with Expanded Cross Neighborhood Re-ranking (2017, background): “In contrast to common neighbors, Jegou et al. [14] use reciprocal neighbors (i.e. common neighbors that reciprocate in a k-neighborhood sense) and propose to compute a contextual dissimilarity measure (CDM).”
Evaluation of GIST descriptors for web-scale image search (2009, methodology): “Another strategy is to use approximate search, as done in [7] by using a two-level inverted file structure or in [2] using min-Hash.”
Improving Bag-of-Features for Large Scale Image Search (2010, unknown): “2007) or using multiple assignment of descriptors to visual words (Jégou et al. 2007; Philbin et al. 2008) at the cost of reduced efficiency.”
Asymmetric metric learning for knowledge transfer (2020, methodology): “In image retrieval , there are efforts to reduce the asymmetry of k -nearest neighbor relations [35], or use asymmetry to mitigate the effect of quantization [33], or handle partial similarity [85] or alignment [67].”
Learning query-dependent prefilters for scalable image retrieval (2009, methodology): “ularies described in [9], which were computed by running the k-means clustering method on a large dataset of Flickr images.”
Bundling features for large scale partial-duplicate web image search (2009, background): “Soft-quantization [4, 13] has been proposed to improve recall by quantizing a descriptor to more than one visual word.”
Unsupervised Person Re-Identification via Multi-Label Classification (2020, methodology): “Label filtering by cycle consistency: Inspired by kreciprocal nearest neighbor[12, 44], we assume that, if two images belong to the same class, their neighbor image sets should also be similar.”
Hamming Embedding and Weak Geometric Consistency for Large Scale Image Search (2008, background): “It is also possible to increase the performance by regularizing the neighborhood structure [10] or using multiple assignment of descriptors to visual words [10, 13] at the cost of reduced efficiency.”
Exploiting the Intrinsic Neighborhood Structure for Source-free Domain Adaptation (2021, background): “Reciprocal neighbors have been studied before in different contexts [14, 31, 60].”
Visual Place Recognition with Repetitive Structures (2015, background): “Scalable place recognition methods [3, 7, 18, 31, 37] often build on the efficient bag-of-visual-words representation developed for object and image retrieval [6, 13, 15, 24, 26, 40].”
Author Manuscript, Published in "ieee Conference on Computer Vision and Pattern Recognition on the Burstiness of Visual Elements (unknown): “In contrast to the method of [5], MA is performed on the query side only.”