Research Activities at Florida State Vision Group

1
Research Activities at Florida State Vision Group

• Xiuwen Liu
• Florida State Vision Group
• Department of Computer Science
• Florida State University
• http//fsvision.cs.fsu.edu

2
Introduction

• An image patch represented by hexadecimals

3
Introduction - continued

• Fundamental problem in computer vision
• Given a matrix of numbers representing an image,
  or a sequence of images, how to generate a
  perceptually meaningful description of the
  matrix?
• An image can be a color image, gray level image,
  or other format such as remote sensing images
• A two-dimensional matrix represents a signal
  image
• A three-dimensional matrix represents a sequence
  of images
• A video sequence is a 3-D matrix
• A movie is also a 3-D matrix

4
Introduction - continued

• Why do we want to work on this problem?
• It is very interesting theoretically
• It involves many disciplines to develop a
  computational model for the problem
• It has many practical applications
• Internet applications
• Movie-making applications
• Military applications

5
Introduction - continued

• How can we characterize all these images
  perceptually?

6
Face Recognition

• Given some examples of faces, identify a person
  under different pose, lighting, and expression
  conditions

7
Face Recognition continued

• Faces of the same person under slightly different
  conditions

8
Affective Computing
9
Face Detection

• Find all faces in a given picture
• Typical faces are available

10
Appearance-based Object Recognition

• Appearance-based object recognition
• Recognize objects based on their appearance in
  images
• Columbia object image library
• It consists of 7,200 images of 100 objects
• Each object has 72 images from different views

11
COIL Dataset
12
3D Recognition Results

• Appearance-based 3D object Recognition
• We compare our result with SVM and SNoW methods
  reported by Yang et al. (Yang et al., 2000)

Methods/Training/test views 36/36 18/54 8/64 4/68
Our method 0.08 0.67 4.67 10.71
Our method without background 0.00 0.13 1.89 7.96
SNoW (Yang et al.,2000) 4.19 7.69 14.87 18.54
Linear SVM (Yang et al.,2000) 3.97 8.70 15.20 21.50
Nearest Neighbor(Yang et al.,2000) 1.50 12.46 20.52 25.37
13
Object Extraction from Remote Sensing Images

• An image of Washington, D.C. area

14
Object Extraction from Remote Sensing Images

• Extracted hydrographic regions

15
Medical Image Analysis

• Medical image analysis
• Spectral histogram can also be used to
  characterize different types of tissues in
  medical images
• Can be used for automated medical image analysis

16
Video Sequence Analysis

• Motion analysis based on correspondence

Video sequence
17
Analytical Probability Models for Spectral
Representation

• Transported generator model (Grenander and
  Srivastava, 2000)
• where
• gis are selected randomly from some generator
  space G
• the weigths ais are i.i.d. standard normal
• the scales ris are i.i.d. uniform on the
  interval 0,L
• the locations zis as samples from a 2D
  homogenous Poisson process, with a uniform
  intensity l, and
• the parameters are assumed to be independent of
  each other

18
Analytical Probability Models - continued

• Define
• Model u by a scaled ?-density

19
Analytical Probability Models - continued
20
Analytical Probability Models - continued
21
Analytical Probability Models - continued
22
3D Model-Based Recognition
23
Summary

• Florida State Vision group offers many
  interesting research topics/projects
• Efficient represent for generic images
• Computational models for object recognition and
  image classification
• Motion/video sequence analysis and modeling
• They can have significant commercial potentials
• They are challenging
• They are interesting

24
Contact Information

• Web site at http//fsvision.fsu.edu
• http//www.cs.fsu.edu/liux
• Email at liux_at_cs.fsu.edu
• Office at MCH 102D
• Office hours Mondays and Wednesdays
  330-530PM
• Phone [phone]
• Courses CAP5615 Fall 2001
• CAP5630 Spring 2001