New Jersey Institute of Technology
Department of Computer Science
CS440 - Computer Vision -
Fall'2008
Wednesday, 6:00 - 9:05 PM, GITC xxx
Course
Description | Readings | Tentative Contents | Grading
Policy | Miscellaneous
Chengjun Liu, Ph.D.
Phone: 973-596-5280
Email: chengjun.liu@njit.edu
Office: GITC
4306
Hours: MW xxxPM-xxxPM or by appointment
Course Description
- This course introduces basic concepts and methodologies of
computer
vision, and focuses on material that is fundamental and has a broad
scope of applications. Topics include contemporary developments in all
mainstream areas of computer vision e.g., Image Fundamentals, Image Formation,
Feature
Representation, Classification and Recognition, Motion Analysis, Camera
Calibration, Stereo Vision, Shape From X (shading, texture, motion,
etc.), and typical applications such as Biometrics.
Readings
- E. Trucco and A. Verri, Introductory
Techniques for 3D Computer Vision, Prentice Hall, 1998.
- R. C. Gonzalez and R. E. Woods, Digital
Image
Processing, 3rd edition, Prentice Hall, 2008.
Tentative
Contents
- Introduction
- Computer Vision Fundamentals
- Related Fields: Image Processing,
Pattern Recognition, Neural Networks, Machine Learning, Artificial
Intelligence
- Image Fundamentals:
- Digital Images: Formats/Protocols -
jpeg,
tiff, png, gif, ppm, pgm
- Color Models/Spaces: RGB, YIQ, YCbCr, YUV, HSV, HSI, XYZ, etc.
- Color Component Images: R, G, B, Y, I,
Q, H, S, V, etc.
- Matlab and Image Processing Toolbox
- 1st Project
- Image Formation
- Basic Optics - Thin Lens
- Basic Radiometry - Image Irradiance and
Scene Radiance, Lambertian Model
- Camera Models - Perspective/Pinhole
Camera Model, Weak-Perspective Camera
Model
- Camera Parameters - Intrinsic and
Extrinsic Parameters
- Improving Image Quality
- Spatial Domain Image Processing - Spatial
Filters
- Frequency Domain
Image Processing - FFT, Convolution Theorem
- Noise Reduction - Lowpass Filtering/Smoothing
- 2nd Project
- Geometric Feature Representation
- Edge Detection - Canny Edge Detector,
Zero-crossing,
LOG, Prewitt, etc.
- Corner Detection - Algorithm
- Line & Curve Detection - Hough
Transform
- Statistical Feature Representation
- Optimal Feature Representation Method
- Principal Component Analysis (PCA)
- Applications Case Study I
- PCA-based Image Compression
- Classification and Recognition Method
- Optimal Feature Classification Method
- Linear Discriminant Analysis (LDA)
- Applications Case Study II
- LDA-based Image Classification
- 3rd Project
- Motion Analysis
- Video Processing Fundamentals -
Component Video, Composite Video, S-Video
- JPEG and MPEG for Image and Video
Compression
- Motion Field and Optical Flow
- Tracking - Kalman Filter
- Structure From Motion
- Biometrics - optional
- Face Recognition
- Iris Recognition
- Fingerprint Recognition
- Camera Calibration - optional
- Intrinsic/Extrinsic Camera Parameters
- Explicit Parameter Calibration
- Projection Matrix-based Parameter Calibration
- 3D Vision - Stereo Vision - optional
- Correspondence
- Epipolar Geometry (E and F matrices)
- 3D Reconstruction
- Shape From X
- optional
- Shape From Shading
- Shape From Texture
Grading
Policy
Projects (topics are related
to our course Contents)
Class presentation/discussion
Class attendance
NJIT Honor Code will be upheld, and any
violations will be brought to the immediate attention of the Dean of
Students.
Students will be consulted with by the
instructor and must agree to any modifications or deviations from the
syllabus throughout the course of the semester.
Miscellaneous