Our goal is to significantly reduce
the computational Star Student Project
load via a distributed visual search Star Student
Project architecture, so that massive reference images
Star Student Project say
millions or billions can be searched based on a group of laptops Star
Student Project or regular servers, or on
cloud computing resources.To establish Star Student
Project a visual search architecture over multiple
servers, we will investigate the distribution of three phases of Star
Student Project local feature
extraction, quantization and inverted indexing,respectively: First,
we Star Student Project
distribute the local feature extraction process by partitioning a
query photo into multiple rectangles, which are Star
Student Project subsequently executed in a
parallel manner