ADAM is a database system to store and retrieve multimedia data. It provides Boolean retrieval and similarity search and makes use of a VA-File indexing for efficient retrieval.
C PLpgSQL C++ Yacc Perl Makefile Other
Clone or download
Fetching latest commit…
Cannot retrieve the latest commit at this time.
Permalink
Failed to load latest commit information.
config
contrib
doc
src
.gitignore
COPYRIGHT
GNUmakefile.in
Makefile
README
README.git
aclocal.m4
configure
configure.in

README

ADAM
=====================================

This directory contains the source code distribution of dbADAM. 

dbADAM is based on PostgreSQL and supports nearest neigbour retrieval for feature vector data. In addition, it comes with Vector Approximation-File to make querying more efficient.

Find more information on dbADAM in the following publications:

- Ivan Giangreco, Ihab Al Kabary, and Heiko Schuldt (2014): ADAM — A Database and Information Retrieval System for Big Multimedia Collections. Proceedings of the 3rd IEEE International Congress on Big Data. Anchorage, USA. 
http://dx.doi.org/10.1109/BigData.Congress.2014.66

- Ivan Giangreco, Ihab Al Kabary, and Heiko Schuldt (2014): ADAM — A System for Jointly Providing IR and Database Queries in Large-Scale Multimedia Retrieval. SIGIR '14: Proceedings of the 37th International ACM SIGIR Conference on Research & Development in Information Retrieval Proceedings. Gold Coast, Australia.
http://doi.acm.org/10.1145/2600428.2611182

ADAM is being developed in the course of the iMotion project (http://imotion-project.eu) at the Databases and Information Systems Research Group (http://dbis.cs.unibas.ch) at the University of Basel, Switzerland. ADAM is no longer in development; the successor ADAMpro can be found at https://github.com/vitrivr/ADAMpro/