ANN: library for approximate nearest neighbor searching

By Sunil Arya, David M. Mount. 1998

Subfields: 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 Information Retrieval & Search

Citations in the corpus

Free-Form Image Inpainting With Gated Convolution (2018, methodology): “To reduce the high cost of memory and computation during search, tree-based acceleration structures of memory [Mount and Arya 1998] and randomized algorithms [Barnes et al. 2009] are proposed.”
Image hallucination with primal sketch priors (2003, background): “This two-level structure can speed up the AAN tree searching algorithm [ 10 ] in the training”
Rank-Based Similarity Search: Reducing the Dimensional Dependence (2015, methodology): “In addition, we compared the performance of RCT against two libraries for approximate k -NN search based on the KD-Tree: the ﬁrst library, ANN [38], provides implementations of the KD-Tree and BD-Tree (box decomposition tree).”
A Hand Motion-guided Articulation and Segmentation Estimation (2020, methodology): “We use kd-tree [23] for searching nearest neighbor points and Voxel Grid Filter to query in the downsampled kd-tree.”
Spatio-Temporal Naive-Bayes Nearest-Neighbor (ST-NBNN) for Skeleton-Based Action Recognition (2017, methodology): “To boost the nearest neighbor searching process, KD-tree implementation [16] is used in our method.”
FAST-LIO2: Fast Direct LiDAR-Inertial Odometry (2021, methodology): “Mapping methods using k-d tree libraries, such as ANN [44], libnabo [43] and FLANN [45], fully re-build the k-d trees to update the map, which results in considerable computation.”
An Efficient k-Means Clustering Algorithm: Analysis and Implementation (2002, methodology): “The data structure implemented was a kd-tree that was constructed from the ANN library [43] using the sliding midpoint rule [39].”
Explicit Approximations of the Gaussian Kernel (2011, methodology): “The dot-dashed line indicates the performance of a 1-NN classiﬁer, trained using the ANN library [Arya and Mount, 1993, Mount and Arya, 2006, Bagon, 2009].”
Efficient Task-Specific Data Valuation for Nearest Neighbor Algorithms (2019, methodology): “Various techniques, such as the kd-tree [43], LSH [15], have been proposed to ﬁnd approximate nearest neighbors.”
Topological Data Analysis (2012, methodology): “The witness graph is easily constructable by computing the | L | × | S | distance matrix using k -nearest neighbors [ 57 ].”
Foreground-Aware Image Inpainting (2019, methodology): “To reduce the runtime and improve memory efficiency, tree-structure based search [21] and randomized methods [5] are proposed.”
Geometric Approximation Algorithms (2011, unknown): “The code of their implementation is available online as a library [AM98].”
Multi-spectral brain tissue segmentation using automatically trained k-Nearest-Neighbor classification (2007, methodology): “Similar experiments were carried out by using rigid registration to MNI305-space. k-Nearest-Neighbor classification The kNN classification was carried out using the ApproximateNearest-Neighbor (ANN) library (www.cs.umd.edu/~mount/ANN/; Mount and Arya, 1998).”
Analysis of approximate nearest neighbor searching with clustered point sets (1999, unknown): “The sliding-midpoint splitting method was first introduced in the ANN library for approximate nearest neighbor searching [MA97].”
Rapid Protein Side-Chain Packing via Tree Decomposition (2005, methodology): “We build SCATD using BALL library [40] for some basic objects such as proteins, residues, and atoms, ANN library [41, 42] to determine if two atoms conflict or not, and split package [43] for tree decomposition of a graph.”