International Joint Conference on AI

IJCAI-2007 Workshop on Multimodal Information Retrieval

Hyderabad, India - January 6, 2007

Final Program
Important Dates
Formatting Guidelines
IBM Research
Supported by IBM Research
IBM Research

Intelligent search for digital information is one of the major challenges of Artificial Intelligence (AI). The many sources of information now available -- text, images, audio, video, time series, sequence data and more -- increase the need for truly multimodal search. Searching for text, images and video is now common in internet search and digital libraries; searching for sequence data is now prevalent in bioinformatics. Clinical records containing unstructured and structured text along with clinical images and video are now being created in healthcare informatics. Finally, time series data is becoming increasingly prevalent in systems management and financial applications.

The fundamental issues in the design of multimodal information retrieval revolve around fast ,accurate selection of data containing an answer to a query. This requires developing robust methods of multimodal feature extraction, data representation, organization, query formulation and search. Further, application-specific retrieval may require exploitation of domain knowledge at all stages of multimodal information retrieval The purpose of this workshop is to bring together researchers from content-based retrieval, AI, database, and application communities who are working in multimodal information retrieval. It calls for original, high-quality submissions that address innovative research and development of multimodal information retrieval systems. Topics of interests include but are not limited to:

  • Content-based indexing, search, and retrieval of multimodal data.
  • Learning and relevance feedback in multimedia retrieval
  • Intelligent agents for multimodal indexing and retrieval
  • Multimodal query and results visualization
  • Multi-modal/multi-sensor fusion techniques
  • Multi-modal event detection and recognition
  • Applications of multimodal retrieval in
    • Bioinformatics
    • Autonomic computing
    • Medical imaging in healthcare.
    • News video.
    • Surveillance
    • Multimodal decision support.
    • Digital libraries
    • Internet search
  • Multimodal database systems and their evaluation.


Papers should not exceed eight pages in length in 12-point font and should remove any author associations to enable double-blind reviewing. All submissions will be peer-reviewed by at least three members of the program committee. Printed as well as electronic proceedings will be available. Please submit your paper at the Workshop submission site.

Formatting Guidelines:

We will be adopting the formatting guidelines of IJCAI with the exception that the number of pages allowed will remain as eight (8). Printed as well as CD-ROM version of the proceedings will be made available to the registered attendees.


  • Deadline for Submission: September 25 2006
  • Notification of Acceptance: October 31 2006
  • Camera-ready Papers Due: November 20, 2006
  • Workshop Date: January 6, 2007


This workshop is intended to introduce the AI community to an important and emerging area at the confluence of multimedia, databases, life science and healthcare informatics. We expect attendees to be members of such communities as well as part of the core AI community interested in developments in this field. With this workshop, we would also like to bring computer vision back to AI.


General Chair:

Narendra Ahuja, Univ. Illinois, Urbana, USA (ahuja at vision dor ai dot uiuc dot edu)

Program Chairs:

Tanveer Syeda-Mahmood, IBM Almaden Research, USA (
Mubarak Shah, Univ. of Central Florida, USA (shah at cs dot ucf dot edu)

Local Organizing Committee:

Santanu Chaudhury, IIT Delhi, India (
C. V. Jawahar, IIIT Hyderabad, India (jawahar at

Program Committee:

Kobus Bernard, Univ. of Arizona,(kobus at cs dot arizona dot edu)
Alberto del-Bimbo, Univ. of Firenze, Italy
Shih-Fu Chang, Columbia University, USA(
Rama Chellappa, Univ. of Maryland, College Park, USA (
Amit Roy-Chowdhury, Univ. California, Riverside, USA (
Rita Cucchiaa, University of Modena and Reggio Emilia, Italy (
Pinar Duygulu-Sahin, Bilkent University, Turkey,
Eric Grimson, M.I.T, USA,
Alan Hanjalic, Delft University of Technology, The Netherlands
Lie LU, Microsoft Research Asia, China(
R. Manmatha, Univ. of Massachusetts, Amherst, USA
Milind Naphade, IBM T.J. Watson Research, USA
PJ Narayan, IIT Hyderabad, India
Raymond Ng, Univ. of British Columbia, Canada
Shin'ichi Satoh, National Institute of Informatics, Japan )
Malcolm Slaney, Yahoo Research, USA (
Arnold Smeulders, Univ. of Amsterdam, Netherlands (
James Wang, Penn. State, USA (jwang AT
Eric Xing, Carnegie Mellon University, USA
Ricardo Baeza-Yaetes, Yahoo Research, Chile
Mohammed Zaki, Rensselaer Polytechnic Institute, USA (
Lei Zhang, Microsoft Research, China


Tanveer Syeda-Mahmood,


Document Retrieval 9:00-10.30
» Indexing and Retrieval of Degraded Handwritten Documents
H.Cao, F. Farooq, and V. Govindaraju
» An Information-theoretic Approach for Automatic Document annotation from Inter- modal Analysis
J. Martinet and S. Satoh
» Content-based Collaborative Filtering Model for Scalable Visual Document Recommendation
S. Boutemedjet and D. Ziou
Machine Learning in Multimodal Retrieval 10.50-12.20
» Face Recognition via Incremental 2D PCA
C. Lu, W. Liu, S. An and Svetha Venkatesh
» Resource-constrained Supervised Dimensionality Reduction
L. Yang, R. Jin, R. Sukthankar
» Semantic image mining via a discriminative model
J. Liu, M. Li, H. Lu, S. Ma, and W. Ma
Lunch 12.30-2.00
Multimodal Retrieval Applications (Invited papers) 2:00-3:00
» Information Mining Challenges in Bioinformatics Data
N. Dimitrova
» Multimodal Mining for Contact Centers: Experiences, Opportunities, and Challenges
G. Pingali, N. Modani, R. Gupta, L. Mignet, T. Syeda-Mahmood, G. Lohman, S. Guven and M. Podlaseck
Multimodal Associations 3:30-4:30
» A Bipartite Graph Model for Associating Images and Text
S. H. Srinivasan and M. Slaney
» Multimodal Retrieval Strategy for Video Collections
A. Malik, A. Jain, M. Matela, S. Chaudhury, S. Bhattacharya