EECE 570: Fundamentals of Visual Computing 

Instructor: Prof. Rafeef Abugharbieh

Fall 2014 Webpage


Calendar Description

Computational and mathematical methods for data driven processing and model-based analysis of digital images and other visual data: perception, capture; representation, modeling; enhancement, restoration; registration, fusion; feature extraction, segmentation; recognition; practical applications.


The course starts with the fundamentals progressing to the state-of-the-art in computational analysis for visual information, i.e. signals/data such as 2D photos, 3D image volumes,  video streams, graphical models etc. The topics draw from a number of exciting fields including image processing, computer vision, shape analysis and geometric modeling, statistical analysis and pattern recognition,  image understanding and artificial intelligence.

Learning Objectives

  • Develop a solid understanding of digital processing and analysis of multi-dimensional image data.
  • Appreciate the basic fundamentals of image capture/acquisition.
  • Understand how to represent such data in the spatial and frequency domains.
  • Perform low level image processing such as denoising, enhancement and restoration.
  • Perform high level image analysis such as segmentation, registration, shape analysis.
  • Get exposure to state-of-the-art techniques in multi-dimensional data computing.
  • Appreciate the recent research developments in the area through research paper reading.
  • Develop experience with related real life practical applications through course projects.
  • Enhance oral and written presentation skills through the interactive learning environment.



  • Elements of visual perception, image acquisition systems.
  • Image representation.
  • Image enhancement in the spatial and frequency domains.
  • Color/vector valued data.
  • Morphological operations.
  • Image restoration.
  • Fundamentals of object localization and representation.


  • Object representation:
    • Region vs. boundary.
    • Explicit vs. implicit models.
    • Statistical models. 
  • Image segmentation:
    • Deformable models.
    • Graph based approaches.
    • Incorporating user guidance and interaction.
    • Probabilistic methods.
  • Image registration and fusion.
  • Motion and deformation analysis .
  • Feature extraction and selection.
  • Object matching and recognition.

      Problem and Project-Based Learning

  • Research paper presentations (written and oral).

  • Individual course project based on real life applications.


Students are expected to be knowledgeable in fundamentals of signal processing and well versed in mathematical preliminaries. Coding skills in MATLAB or another suitable platform/language is essential for project work.

Course Structure

The course material will be presented through a combination of lectures, group readings and discussions, practice-based learning and term course project.

Evaluation scheme

Students will be evaluated based on attendance, active participation, homework assignments, reports, in class presentations and discussions of research papers, and a (significant) term course project.