Retrieval of Web Images for Computer Vision Research

This site publishes the Bachelor's Thesis Retrieval of Web Images for Computer Vision Research written by Maikel Linke and supervised by Marco Kortkamp and Sven Wachsmuth. It was written at the Applied Informatics Group at the Faculty of Technology, Bielefeld University.

Abstract

Today, the web is an important source of image data for computer vision research. Search engines and online photo albums allow the retrieval of images related to a given keyword. This work contains an overview of these services and their usage. Additionally, it describes the development of a common programming interface to access different services. The developed Java library implements the most prominent services.

Download the thesis as PDF. You can use this BibTeX key to cite. If you are interested in the LaTeX source, just click.

Java library with source code

The developed Java library is also available:

• WebImageRetrieval.tar.gz, with source.
• WebImageRetrieval.zip, with source.

These archives contain the following files among others:

• WebImageRetrieval.jar depending on
• flickrapi-1.2.jar, the Flickr API.
• Browse WebImageRetrieval JavaDoc.

Warning: You can use this software on your own risk. You have to consider the terms of usage of each used service (Bing, Flickr, Google). There will be no official support for this software.