This project implements large-scale image matching using Hadoop and Pig.
Web: A Javascript-based data mining tool for downloading pictures from Google StreetView.
Pig: Pig Latin scripts for constructing a bag-of-features database from a list of image feature descriptors, and performing queries on the database with an input image.
Py: Various python scripts for data mining and image processing
Modules: An implementation of Locality Sensitive Hashing for the Hamming distance, as a Python module in C++. Install script is included.