Concise Computer Vision
- A One-semester Course -
A 4th-year course in Computer Science taught by Professor Reinhard Klette
in the second semester 2014 at Tamaki campus, The University of Auckland, New Zealand (COMPSCI 775).
The course consists of 30 lectures (50 min each), 8 tutorials,
three assignments, one supervised camera calibration and recording session with HAKA1, 3 seminars (for student presentations of assignment results), and
one review of all the taught material prior to the exam date.
The lectures follow material published in the
book Concise Computer Vision by Reinhard Klette (January 2014, Springer, London, UK). For example, Amazon sells the Kindle version for less than 50 US-dollars.
Tutors: Dongwei Liu and Haokun Geng,
PhD students in the
.enpeda.. project.
The course assignments count 30% of the total mark, and the exam 70%.
There will be three individual assignments using video data recorded with cameras
installed in HAKA1, the test vehicle of the .enpeda.. group:
(1) Computational photography
(2) Semantic segmentation
(3) Stereo vision and visual odometry
For each assignment, a report needs to be submitted by the due
date (as announced in lectures). Solutions and results may be presented in
seminars. If you like to use Latex for writing your report then
you can start with this template.
LECTURES or SEMINARS
TUE 11:30am - 12:30pm (733-231) and THU 10:30am - 12:30pm (723-231)
TUTORIALS or LECTURES
FRI 11:30am - 12:30pm (721-234)
RECORDING (with cameras installed in test-vehicle HAKA1)
by individual arrangement with the tutors
Schedule
- 22 July (1): Course Intro
- 24 July (2):
Images and Fourier
- 25 July (1): Color
29 July: Tutorial (OpenCV)
31 July: Tutorials (OpenCV and Assignment 1)
-
5 August (1): Operators
- 7 August (2): Smoothing, Sharpening, Edges (Basics) and Corners and Edges
- 8 August (1): Image Segmentation 1
- 12 August (1): Image Segmentation 2
- 14 August (2): Image Segmentation 3 and Adjacency, Regions, Borders
15 August: Tutorial (Assignment 2)
18 August: Deadline for Assignment 1 (12 noon)
- 19 August (1): Area, Length, Curvature
- 19 August (2): Image Value Analysis; Line and Circle Detection and Cameras
- 19 August (2): Coordinates and Calibration and
Stereo Vision and Rectification
- 20 August (SEMINAR) Computational Photography and (1): Stereo Matching - Basics
15 September: Deadline for Assignment 2 (12 noon)
16 September (HAKA1 Intro): Tutorial
- 23 September (1): Pixel Labeling: Stereo Vision
- 25 September (2): Pixel Labeling: Optic Flow
and Tutorial (Assignment 3, Stereo Vision and Visual Odometry)
- 30 September (1): Pixel Labeling: Photometric Stereo
- 2 October (2): Image Features 1 and Image Features 2
Recording of sequences in HAKA1 (arrange time with tutor)
- 7 October (1): (SEMINAR) Semantic segmentation
- 9 October (2): Feature Tracking and Particle Filter
- 14 October (1): Kalman Filter
- 16 October (2): Object Detection 1
and Object Detection 2
20 October: Deadline for Assignment 3 (12 noon)
- 21 October: (1)
(SEMINAR): Stereo Vision and Visual Odometry and, if time allows, Object Detection 3
- 23 October (2): Vision-based driver assistance
(for applying or recalling concepts discussed in the lectures)
and Questions (examinable only if discussed in lectures; there
are additional questions in this file having bold numbers)
Copyright for material (above) on this website (if not otherwise specified): Reinhard Klette
DATA LINKS
A few images (as used in the CCV book)
Image textures
Video sequences on EISATS
FURTHER MATERIAL
Brief intro into OpenCV by Marvin Smith
and there are many more OpenCV introductions online
FREE SYSTEMS or SOURCES
GIMP and
ImageJ
See also (older) software-links on
The Computer Vision Homepage (only updated till 2006).