Thick boundaries in binary space and their influence on nearest-neighbor search
► We show that binary-space Voronoi diagrams have very thick boundaries. ► This leads to a performance loss of the binary nearest neighbor search algorithms. ► We present two effective ways to overcome this limitation. ► They are based on randomized data partitioning. Binary descriptors allow faster...
Uloženo v:
| Vydáno v: | Pattern recognition letters Ročník 33; číslo 16; s. 2173 - 2180 |
|---|---|
| Hlavní autoři: | , , |
| Médium: | Journal Article |
| Jazyk: | angličtina |
| Vydáno: |
Elsevier B.V
01.12.2012
|
| Témata: | |
| ISSN: | 0167-8655, 1872-7344 |
| On-line přístup: | Získat plný text |
| Tagy: |
Přidat tag
Žádné tagy, Buďte první, kdo vytvoří štítek k tomuto záznamu!
|
| Shrnutí: | ► We show that binary-space Voronoi diagrams have very thick boundaries. ► This leads to a performance loss of the binary nearest neighbor search algorithms. ► We present two effective ways to overcome this limitation. ► They are based on randomized data partitioning.
Binary descriptors allow faster similarity computation than real-valued ones while requiring much less storage. As a result, many algorithms have recently been proposed to binarize floating-point descriptors so that they can be searched for quickly. Unfortunately, even if the similarity between vectors can be computed fast, exhaustive linear search remains impractical for truly large databases and approximate nearest neighbor (ANN) search is still required. It is therefore surprising that relatively little attention has been paid to the efficiency of ANN algorithms on binary vectors and this is the focus of this paper.
We first show that binary-space Voronoi diagrams have thick boundaries, meaning that there are many points that lie at the same distance from two random points. This violates the implicit assumption made by most ANN algorithms that points can be neatly assigned to clusters centered around a set of cluster centers. As a result, state-of-the-art algorithms that can operate on binary vectors exhibit much lower performance than those that work with floating point ones.
The above analysis is the first contribution of the paper. The second one is two effective ways to overcome this limitation, by appropriately randomizing either a tree-based algorithm or hashing-based one. In both cases, we show that we obtain precision/recall curves that are similar to those than can be obtained using floating point number calculation, but at much reduced computational cost. |
|---|---|
| ISSN: | 0167-8655 1872-7344 |
| DOI: | 10.1016/j.patrec.2012.08.006 |