Digital image processing Computer Vision

1
Digital image processingComputer Vision

• Instructor
• Asst. Prof. Dr. Nualsawat Hiransakolwong
• ??. ??. ??????? ????????????
• ?. ??? ???. 0814057819

2
Overview

• Introductory computer vision course in the
  Computational Perception and Robotics (CPR)
  curriculum
• Course web site
• www.kmitl.ac.th/khnualsa/doc/cv.htm

3
Course Objectives

• Broad introduction to research problems in
  computer vision, for example
• How to recover depth information?
• How to recognize objects?
• Familiarity with standard vision results, such
  as
• Multi-scale image representations
• Finding correspondences between images
• Experience with programming vision algorithms
• MATLAB
• C

4
Prerequisites

• Linear algebra
• Least squares problems
• Singular Value Decomposition
• Programming experience
• MATLAB a help but not required
• One or more MATLAB/linear algebra tutorial
  sessions will be offered

5
Logistics

• Text Computer Vision by Forsythe and Ponce
• Should be in bookstore now

6
Workload

• This course is going to be a lot of fun.
• Its also going to be some amount of work.
• If you dont have programming experience, then
  its going to be a lot of work.
• If you have questions or concerns, dont hesitate
  to send me email or call me.

7
The World in an Eye
8
Jitter Camera
9
Synthetic aperture imaging
Cool QuickTime movie
10
Multi-Camera Array
11
The Lightfield (Gibsons optic array)

• Plenoptic Function
• 5DOF wavelength time
• image 2D slice
• Do not throw away samples from the lightfield !

12
Plenoptic Sampling
1
N
Space
1
Snapshot
Stereo, Multiview
Surveillance
T
Tracking
Time
13
Unstructured Road Following
14
Video Google
Query frame with outlined query region and its
close-up
Retrieved key-frames from three different shots
15
Visual Odometry
16
Wide-baseline Stereo
17
3D Object Recognition
18
Active Appearance Models
19
Video Matting
20
Tracking Ants
21
Honeybees
22
Tracking Articulated Shapes
23
Tracking Loose-Limbed people
24
Tracking Multiple Humans
25
Viola-Jones Face Detector
26
Geometric Context from a Single Image
Hoiem, Efros, Hebert
27
Automatic Pop-up
28
Putting Objects in Perspective