Assignment 1 - Experiment & Review

Due: 14.3.2005

1.      Search for a content-based retrieval system on the Web (from 2000 on).

Keywords:  content-based, image indexing, image retrieval, medical image retrieval.

Example of earlier systems are provided below.
 
 

IBM: 
Qbic

System: http://wwwqbic.almaden.ibm.com

Berkeley: 
Blobworld

System: http://elib.cs.berkeley.edu/photos/blobworld/
Biblio:
Chad Carson, Serge Belongie, Hayit Greenspan, and Jitendra Malik, Blobworld: Image segmentation using Expectation-Maximization and its application to image querying, Third International Conference on Visual Information Systems 1999. [download]

MIT: 
Photobook

System: http://www.media.mit.edu/~tpminka/photobook/
Biblio:
A. Pentland, R. Picard, and S. Sclaroff, Tools for Content-Based Manipulation of Image Databases, SPIE Storage and Retrieval of Image & Video Databases II, Feb 1994.  [download]

Colombia: 
VisualSEEk 

System: http://disney.ctr.columbia.edu/SaFe/
Biblio:
J. R. Smith and S.-F. Chang, VisualSEEk: A Fully Automated Content-Based Image Query System, ACM Multimedia Conference, Boston, MA, Nov. 1996. [download]

Santa Barbara: 
NETRA

System:  http://maya.ece.ucsb.edu/Netra
Biblio:
W. Y. Ma and B. S. Manjunath, " NETRA: A toolbox for navigating large image databases," in IEEE International Conference on Image Processing, 1997, and ACM Multimedia Systems journal  [download]

Stanford:

System:  http://robotics.stanford.edu/~rubner/
BIblio:
Y. Rubner, C. Tomasi, and L. J. Guibas. A Metric for Distributions with Applications to Image Databases. Proceedings of the 1998 IEEE International Conference on Computer Vision, Bombay, India, January 1998. [download]

Rutgers: 
Image Guided Decision Support (IGDS)

System: http://www.caip.rutgers.edu/~comanici/jretrieval.html
Biblio:
D. Comaniciu, P. Meer and D. Foran, Image Guided Decision Support (IGDS) System, submitted to Machine Vision and Applications 1998 [download]


 
 

2.

Read 1-2 articles or technical reports that describe the chosen system.

3.

Write a short review, 1-2 pages text (strict!) + queries picture examples, about the system. In the review, address the following issues (those that are relevant to the chosen system): 

 

a.

Describe input databases available in the system (content/size, etc..) & its flexibility to accept new, user defined, databases.

 

b.

Describe the query types possible in the system - 

 

 

-
-
-

What feature types? (color, texture, shape, etc... )
Can the user combine features? 
Can the user provide preferences (e.g. prioritizing features)?

 

c. 

Methods and algorithms used for indexing (no need to go into much detail)

 

d. 

(**)The systems are available online or for download 
(QBIC – download to PC, Photobook – sources available to be ftped to UNIX; Rutgers - read only) 
Experiment with querying the system - you are now the User: 

 

 

-
-
-

Describe several examples you were satisfied with.
Describe examples you considered as "wrong".
Do you have special requests for improvement?


 
 

Interesting summaries will be chosen to be presented in the class in 10-15 minute presentations