Efficient Large-scale Approximate Nearest Neighbor Search on the GPU (Product-Quantization-Tree)

Patrick Wieschollek (Universität Tübingen, Max-Planck-Institut für Intelligente Systeme)
Hendrik Lensch (Universität Tübingen)
Oliver Wang (Adobe)
Alexander Sorkine-Hornung (Disney Research Zurich)

IEEE Conference on Computer Vision and Pattern Recognition (CVPR) 2016

June 26, 2016

paper.pdf
supplemental.pdf

We present a new approach for efficient approximate nearest neighbor (ANN) search in high dimensional spaces, extending the idea of Product Quantization. We propose a two level product and vector quantization tree that reduces the number of vector comparisons required during tree traversal.

Our approach also includes a novel highly parallelizable re-ranking method for candidate vectors by efficiently reusing already computed intermediate values. Due to its small memory footprint during traversal the method lends itself to an efficient, parallel GPU implementation.

This Product Quantization Tree (PQT) approach significantly outperforms recent state of the art methods for high dimensional nearest neighbor queries on standard reference datasets.
Ours is the first work that demonstrates GPU performance superior to CPU performance on high dimensional, large scale ANN problems in time-critical real-world applications, like loop-closing in videos.

@inproceedings{PQT,
    author = "Patrick Wieschollek and Oliver Wang and Alexander Sorkine-Hornung and Hendrik P.A. Lensch",
    title = "Efficient Large-scale Approximate Nearest Neighbor Search on the GPU",
    booktitle = "IEEE Conference on Computer Vision and Pattern Recognition (CVPR)",
    pages = "",
    month = "June",
    year = "2016"
}