Summer School on

Manifold Learning in Image and Signal Analysis

Backfallsbyen, Ven, Sweden, August 17 - 21, 2009 (both days included)

Summer School on Manifold Learning in Image and Signal Analysis

Many interesting problems involve the description of and statistics on hyper-dimensional manifolds. An example is biological shape, where a single shape can be represented as a point in a high-dimensional shape space. A set of shapes will therefore constitute a sampled manifold in that space. The topic of this summer school is the mathematical and statistical description of such manifolds. An important aspect is the ability to reduce the dimensions of the embedding space. Typically, this is done by removing the dimensions that contain redundant information or noise.

Topics:

  • Geometry and representation of high dimensional non-linear manifolds
  • Manifold description of biological shape, human motion analysis, and other natural phenomena
  • Dimensionality reduction using:
    • Principal Component Analysis (PCA)
    • Sparse-PCA
    • Kernel-PCA
    • Independent Component Analysis (ICA)
    • Locally linear embedding (LLE)
    • Isometric mapping (ISOMAP)
    • Laplacian Eigenmaps
    • Diffusion maps

The course will consist of lectures given by members of the research groups at IMM-DTU and DIKU as well as invited speakers with expertise in the field of non-linear dimensionality reduction.

Last minute information

Article (in Danish) about the summer school
Welcome letter
Map to the boat
House distribution

DSAGM

The annual meeting of the Danish Association For Automatic Recognition of Patterns (DSAGM) (Dansk Selskab for Automatisk Genkendelse af Mønstre) will be held together with the summer schools. Further information can be found on the DSAGM page and here.

Credits

Ph.D. ECTS: You get 2.5 ECTS from just participating. You can get additionally 2.5 ECTS if you make a poster or oral presentation during the summerschool. Finally, you can get additionally 2.5 ECTS (if you did not do a poster or oral presentation) by writting a small report on how the topics covered on the course relate to your own phd project. That is, you can get in total 5 ECTS for this course.

Organisation:

The summer school is organised in collaboration between the Image Group, Department of Computer Science, University of Copenhagen, the Image Analysis and Computer Graphics section, DTU Informatics, Technical University of Denmark, the ITMAN graduate school programme at DTU Informatics, and the Graduate School of Science University of Copenhagen.


DSAGM


Registration and practical matters

NEW: The summer school is fully booked.

Summer School on Manifold Learning in Image and Signal Analysis
Backfallsbyen, Ven, Sweden, August 17 - 21, 2009 (both days included)
Dina Riis Johannessen dinariis@diku.dk