Approximately Nearest Neighborhood Image Search Using Unsupervised Hashing via Homogeneous Kernels

Table 2

Performance comparison based on the Caltech-101 data set.

Number of training samples	Proposed method	[25]	[29]	[30]	[24]	[31]	[32]
15		59.1	56.4	52	51	49.5	44
30		66.2	64.6	N/A	56	58.2	63