University of Tsukuba | Grad. Scho. Syst. and Info. Eng. | Dept. Comp. Sci. | List of Courses
Keisuke Kameyama
URL (guest access allowed)
Office hours Please contact by mail.
Cource# 01CH609, 01CF114
Area Media Engineering
Course style Lecture
Term Spring AB
Period Mon2
Room# 3B303
Keywords Pattern Recognition, Adaptation, Feature Extraction, Image Processing, Machine Learning
Prerequisites Basic knowledge on Linear Algebra, Analysis, Probability and Statistics of undergraduate level. Understanding of basic signal processing would be a plus.
relation degree program competence Knowledge Utilization Skills,International Skills,Research Skills,Expert Knowledge
Outline Adaptive techniques in processing, recognition and retrieval of media information will be discussed. Much weight will be put on (re-)assuring the fundamental knowledge and algorithms in machine learning and signal/image processing, that are essential for adaptive handling of media contents. In addition, up-to-date methods in the field will also be mentioned.
Course plan Weeks 1-2
Introduction and reviews on math used in this course.

Weeks 3-7
Theories and techniques for adaptation, recognition and retrieval.
  • Basic Pattern Recognition and the Bayes Rule
  • Linear Discrimination and Adaptive Filters
  • Neural Networks
  • Support Vector Machines
  • Clustering
  • Nearest Neighbor and Subspace Methods

  • Weeks 8-10
  • Content-Based Image Retrieval (CBIR)
  • Biometric Authentication
  • Classification of general object images
  • Textbook
  • C. Bishop, Neural networks for pattern recognition, Oxford Univ. Press 1995
  • Haykin, Neural networks - A comprehensive foundation - Prentice Hall 1998
  • F. M. Ham, I. Kostanic, Principles of neurocomputing for science and engineering, McGraw-Hill, 2001
  • C. Bishop, Pattern recognition and machine learning, Springer 2006 (邦訳あり)
  • 熊沢逸夫、学習とニューラルネットワーク、森北出版.
  • Evaluation Irregular small assignments and final term paper will be evaluated.
    TF / TA None
    Misc. Lecture in English