2012 IEEE Conference on Computer Vision and Pattern Recognition (2012)
Providence, RI USA
June 16, 2012 to June 21, 2012
The k-NN graph has played a central role in increasingly popular data-driven techniques for various learning and vision tasks; yet, finding an efficient and effective way to construct k-NN graphs remains a challenge, especially for large-scale high-dimensional data. In this paper, we propose a new approach to construct approximate k-NN graphs with emphasis in: efficiency and accuracy. We hierarchically and randomly divide the data points into subsets and build an exact neighborhood graph over each subset, achieving a base approximate neighborhood graph; we then repeat this process for several times to generate multiple neighborhood graphs, which are combined to yield a more accurate approximate neighborhood graph. Furthermore, we propose a neighborhood propagation scheme to further enhance the accuracy. We show both theoretical and empirical accuracy and efficiency of our approach to k-NN graph construction and demonstrate significant speed-up in dealing with large scale visual data.
learning (artificial intelligence), approximation theory, computer vision, graph theory, large scale visual data, visual dcscriptors, data-driven techniques, learning tasks, vision tasks, large-scale high-dimensional data, approximate k-NN graphs construction, data points, base approximate neighborhood graph, Accuracy, Complexity theory, Approximation algorithms, Indexing, Nearest neighbor searches, Visualization
Shipeng Li, Rui Gan, Zhuowen Tu, Gang Zeng, Jingdong Wang and Jing Wang, "Scalable k-NN graph construction for visual descriptors," 2012 IEEE Conference on Computer Vision and Pattern Recognition(CVPR), Providence, RI USA, 2012, pp. 1106-1113.