Approximate Nearest Neighbors (NN) Libraries in high-dimensions | RKD-forest vs LSH vs BBD-tree


The RKD forest is a project we created. For LSH, we used E²LSH (from MIT) and for the BBD tree we used ANN library. ANN was easier to handle than E²LSH. We used SIFT data for the measurements.

BBD tree was the most accurate of all, while RKD forest the fastest of all. BBD couldn’t decrease much of its execution time by approximation parameters (such as epsilon) will E²LSH could, resulting in faster execution with just a bit loss in precision (in respect with BBD tree).


Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Google+ photo

You are commenting using your Google+ account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )


Connecting to %s