TY - GEN
T1 - Scalability of the NV-tree: Three Experiments
AU - Amsaleg, Laurent
AU - Jónsson, Björn Thór
AU - Lejsek, Herwig
PY - 2018/10
Y1 - 2018/10
N2 - The NV-tree is a scalable approximate high-dimensional indexing method specifically designed for large-scale visual instance search. In this paper, we report on three experiments designed to evaluate the performance of the NV-tree. Two of these experiments embed standard benchmarks within collections of up to 28.5 billion features, representing the largest single-server collection ever reported in the literature. The results show that indeed the NV-tree performs very well for visual instance search applications over large-scale collections.
AB - The NV-tree is a scalable approximate high-dimensional indexing method specifically designed for large-scale visual instance search. In this paper, we report on three experiments designed to evaluate the performance of the NV-tree. Two of these experiments embed standard benchmarks within collections of up to 28.5 billion features, representing the largest single-server collection ever reported in the literature. The results show that indeed the NV-tree performs very well for visual instance search applications over large-scale collections.
KW - High-dimensional indexing
KW - Visual instance search
KW - Scalable methods
KW - Large-scale collections
KW - Performance evaluation
KW - High-dimensional indexing
KW - Visual instance search
KW - Scalable methods
KW - Large-scale collections
KW - Performance evaluation
U2 - 10.1007/978-3-030-02224-2_5
DO - 10.1007/978-3-030-02224-2_5
M3 - Article in proceedings
T3 - Lecture Notes in Computer Science
SP - 59
EP - 72
BT - Proceedings of the International Conference on Similarity Search and Applications (SISAP)
A2 - Marchand-Maillet, Stéphane
A2 - Silva, Yasin N.
A2 - Chávez, Edgar
PB - Springer
CY - Lima, Peru
ER -