ICML Workshop on

Machine Learning in Computer Vision

July 9, 2002
Sydney,  Australia
In conjunction with ICML-2002 The Nineteenth International Conference on Machine Learning
(text format)

Workshop description

Learning is one of the current frontiers for computer vision research and has been receiving increased attention in recent years. Machine learning technology has strong potential to contribute to:
        - the development of flexible and robust vision algorithms that will improve the performance of practical vision systems with a higher level of  competence and greater generality, and
        - the development of architectures that will speed up system development time and provide better performance.

The goal of improving the performance of computer vision systems has brought new challenges to the field of machine learning, for example, learning from structured descriptions, partial information, incremental learning, focusing attention or learning regions of interests (ROI), learning with many classes. Solving problems in visual domains will result in the development of new, more robust machine learning algorithms that will be able to work in more realistic settings.

From the standpoint of computer vision systems, machine learning can offer effective methods for automating the acquisition of visual models, adapting task parameters and representation, transforming signals to symbols, building trainable image processing systems, focusing attention on target object and learning when to apply what algorithm in a vision system.

From the standpoint of machine learning systems, computer vision can provide interesting and challenging problems for example: learning models rather than hand-crafting them, learning to transfer experience gained in one application domain to another domain, learning from large sets of images with no annotation, designing evaluation criteria for the quality of learning processes in computer vision systems.

Workshop schedule

9:15 - 9:30am Welcome and registration -
9:30 - 10:00am Paper 1: 
Recent Progress on RAIL: Automating Clustering and Comparison of Multiple Classification Techniques on High Resolution Remotely Sensed Imagery [pdf]
A. Chen
School of Computer Science and Engineering, University of New South Wales
10:00 - 10:30am Paper 2:
Boosted Image Classification: An Empirical Study [pdf]

N. R. Howe
Smith College
10:30 - 11:00am Coffee break -
11:00 - 11:30am Paper 3:
Signal Discrimination in Fluorescence In Situ Hybridization Images [pdf]

B. Lerner
Department of Electrical and Computer Engineering, Ben-Gurion University
11:30 - 12:00pm Paper 4:
Combining Wrapper and Filter Approaches for Learning Concepts from Images provided by a Mobile Robot [pdf]
N. Bredeche
12:00 - 2:00pm Lunch break -
2:00 - 2:30pm Paper 5:
Automatic Feature Construction and a Simple Rule Induction Algorithm for Skin Detection [pdf]
ITESM - Campus Cuernavaca
2:30 - 3:00pm Paper 6:
A Statistical Approach To Texture Description of Medical Images: A Preliminary Study [pdf]
M. Bevk
Faculty of Computer and Information Science, University of Ljubljana
3:00 - 3:30pm Paper 7:Learning to Recognize Objects - Toward Automatic Calibration of Colour Vision 
for Sony Robots [pdf]
T. Zrimec
Centre for Health Informatics & SCE
University of New South Wales  
3:30 - 4:00pm Coffee break -
4:00 - 5:00pm Panel discussion -


Program co-chair

    Arcot Sowmya
    University of New South Wales
    E-mail  sowmya@cse.unsw.edu.au

    Tatjana Zrimec
    University of New South Wales
    E-mail   tatjana@cse.unsw.edu.au

Additional information

For additional information, see the web site for the conference: