Introduction

This set of web pages was designed to give prospective students of Florida State Vision Group an overview of the research activities happening at the group, research assistantship opportunities, and a list of master projects, master thesis topics, and Ph.D. dissertation topics for Computer Science students at Florida State University.

Research activities at Florida State Vision group are centered around visual information processsing from human visual perception modeling, computational models and algorithms for perception problems, to real world applications. Visual information processing is a very complex phenomenon that is involved with prior knowledge stored in memory, moldeing of the visual environment, and computation at a given moment of deriving a solution for the purpose of survival and decision making.

For example, if you look at the image on the right, without any efforts, you see that someone is playing tennis. However, it turns out to be a very difficult problem if you are asked to design a computer program which can produce similar output given the input (By the way, you think that you know how to solve this problem, please let me know and I will help you to become the richest guy in the world). We believe that the central question behind the vision problem is a modeling problem. That is, could we come up with a computation model whose input-output behavior is similar to ours?

Because our visual information processing is a complex system, scientists and engineers in working this area attempt to decompose the system into sub-systems with the hope that the sub-systems are easier to solve. Then by integrating the sub-systems together, we could have an approximate system which would have the most useful input-output relationships for real-world applications.

Research Topics

As discussed in the introduction, our research activities are centered around vision problems. Based on your interest emphasis and experience, our group provides a wide range of problems that are suitable as Master projects, Master thesis topics, and Ph.D. dissertation topics.

Novel ideas are always welcome here at Florida State Vision group. If you have a novel idea on solving a problem, please feel free to discuss with the group members and generous support will be provided if the idea fits to the group research agenda.

Master Projects

• Face detection based on spectral histograms
  This project is to explore how to use spectral histogram representation to find faces in different kinds of images.
• Content-Based Image Retrieval
  This project is to explore how to use spectral histogram representation to find images that contain similar objects of interest in a large image database.

Master Thesis Topics

• Human-Computer Interactions
  This topic is to develop effective representation and efficient algorithms for the purpose of human-computer interactions.
• Image-Based Computer Graphics
  This topic is to explore the techniques for building realistic models for computer graphics purpose. An example would be to build a face model with realistic facial expression from a sequence of images.
• Feature Extraction from Remote Sensing Images
  This topic is to explore the automatic extraction of useful information from remote sensing images, including satellite images and aerial images.
• Medical Image Analysis
  This topic is to explore the automatic diagnosis of abnormality based on images from different sources.

Ph.D. Dissertation Topics

• Motion Segmentation
  This topic is to explore effective and efficient techniques to derive visual information from a sequence of images. This involves modeling, computation, and implementation of motion analysis.

Research Assistantship Opportunities

The Florida State Vision Group has been fortunate in obtaining fundings to support our search over the years. The secret of our group's success is through superior performance achieved through novel ideas. In this regard, our group sponsors well motivated Ph.D. students interested in vision and related real-world applications with generous resource supports.

Last modified on April 15, 2001