Acoi

The Acoi database is a Dutch national experimentation platform for multimedia storage, indexing and retrieval. The index is geared at providing access to one million still images, hundreds of video sequences, and thousands of audio tracks. Its novel plug-in architecture provides the means to subsequently index these multimedia objects using feature extraction algorithms published in the literature or developed by the partners in the various affiliated research projects.

Advanced Query and Retrieval Techniques for Remote Sensing Image Databases (RSIA II+III)

Search for Satellite Images of Switzerland with Image Content Search Search for Image Content using spectral and textural properties. Search for signal classes directly, derived rules, or pre-defined cover-types.

Amore - NEC

Several implementations (web interfaces) are available for searching in this database. The user can not influence the results of the search very much in the web interface.

Artisan

ARTISAN is a shape retrieval system designed to handle the special needs of trademark registration and retrieval.

Blobworld

Blobworld is a retrieval system based on blobs. While Blobworld does not exist completely in the "thing" domain, it recognizes the nature of images as combinations of objects, and querying and learning in Blobworld are more meaningful than they are with simple "stuff" representations. An interactive demo is on line.

Candid

Research at Los Alamos National Laboratory has resulted in the development of CANDID, the Comparison Algorithm for Navigating Digital Image Databases. This algorithm facilitates content-based retrieval of digital imagery from large databases using a query-by-example methodology. A user provides an example image to the system, and images in the database that are similar to that example are retrieved.

Chabot

For retrieval, Chabot uses tools provided by POSTGRES, such as representation of complex data types, a rich query language, and extensible types and functions. This project is finished.

Compass

COMPASS is a distributed application for content-based image retrieval using remote databases. The COMPASS system can be used for two main activities: to browse still image databases, and to search image databases similar to a query image. Therefore, the user formulates queries by examples and not a mere caption textual search

Content-Based Image Retrieval for Medical Image

To pose a query to the database, the physician circles one or more pathology bearing regions (PBR) in the query image. The system then retrieves the n most visually similar images from the database using an index comprised of a combination of localized features of the PBRs and of the global image. Within this framework we are investigating several computer vision and machine learning research issues, which include characterizing medical images, efficient retrieval, and customized queries. The system Assert is an on line demo.

Excalibur Visual Retrieval Ware

The company is experienced in searching in text databases. They have an implementation of an image search engine and examples on line. Users can search on many combinations of index parameters, and furthermore they have a SDK for implementing new function. A video search engine is also available. This software is also implemented by Yahoo search engine.

Image Database Retrieval of Rotated Objects by User Sketch

This demo shows an image retrieval system, which enables the user to search a greyscale image database intuitively by presenting simple sketches. The database contains 120 different isolated objects (mostly hand tools) with arbitrary orientation. Each of these images is represented by a Hidden Markov Model (HMM) which has been modified in order to obtain rotation and scale invariance. Thus, no labeling of the content or the rotation angle of the object is needed when adding new images to the database.

ImageScape

Sketch based search engine for images on WWW (in Java). University of Leiden.

Image Retrieval and Database Navigation Demo

Imatch

This software package for Windows environment is convenient to use in image databases. It is easy to install and does not cost as much as other commercial packages, whereas it has some good retrieval algorithms implemented.

Metaseek

MetaSEEk is designed to intelligently select and interface with multiple on-line image search engines by ranking their performance for different classes of user queries. User feedback is also integrated in the ranking refinement. An on line demo is available.

MPEG-7

The development of this standard which is for streaming video, is believed to have a high impact on new search algorithms in image databases. A web version is on line.

Netra

NETRA is a prototype image retrieval system that is currently being developed in the UCSB Alexandria Digital Library (ADL) project. NETRA uses color, texture, shape and spatial location information in segmented image regions to search and retrieve similar regions from the database. There is a java demo available. There is also a Netra-2 version that is more sophisticated in color image segmentation and local color feature.

Photobook - MIT

For use in research this database has source codes available and users can implement there own code in this database. For compiling these databases users have to be experienced in compiling c-sources.  There are also nice examples available on line. In version 5, these include euclidean, mahalanobis, divergence, vector space angle, histogram, Fourier peak, and wavelet tree distances, as well as any linear combination of these. Version 6 allows user-defined matching algorithms via dynamic code loading.

QBIC Image database of IBM

This database is easy to use for searching images. It has a web interface, and new images can be added easily. The software can be downloaded on line. Implementing other algorithms for searching is difficult. An example of images of logo's on pills that I have implemented is available at http://forensic.to/drugspills .

Search by content

This page presents some of the activity of the visal computing lab at UCSD in image retrieval by psychologically based similarity. With a demo.

Simplicity

Semantics-sensitive Integrated Matching for Picture LIbraries; on line demo in java.

Surfimage

An on line web demo of INRIA with relevance feedback.

Vicar

VICAR aims at the development of a system for the documentation and analysis of digital video material. The project result will combine a powerful video browsing and annotation editing program with a suite of software tools for the automatic extraction of semantic contents from MPEG 1 and MPEG 2 video streams.

Virage

The company advertises that they can index video sources. They have several examples on line. Furthermore they have several patents on the search engine they use. It is a very promising way of searching video streams on line. The image search engine is included in Oracle 8.

Visualseek

VisualSEEk is a Web tool for searching for images and videos. VisualSEEk allows the user to make queries using visual features. The demo system currently supplies 3,200 (and now also 12,000) color images and videos.

Zomax Pic2Seek

This database is Java-based and can be used on line. It is a huge database of images, and users can search on structures and colour (invariant to light conditions). All images that are available on the www can be entered in this database. The source is not available on line.