ENGG5104 Image Processing and Computer Vision
(2016 - 2017)

[ Course Info | Lecture Notes | Projects ]

 

Course overview:

It is getting easier and more common to acquire an image in digital form using advanced digital image acquisition devices, such as digital mobile phones, digital cameras, video recorders, CCD, finger print scanners, and satellites. As the use of these devices is widespread, the need of processing, understanding and compressing a digital image is growing particularly in electronic security surveillance technology in public spaces (e.g., recognition and motion tracking), diagnosis in medical images, temperature and weather monitoring and analysis in satellite images.  

The goal of computer vision is to compute properties of the three-dimensional world from digital images. Problems in this field include identifying the 3D shape of an environment, determining how things are moving, and recognizing familiar people and objects, all through analysis of images and video. This course provides an introduction to computer vision, including feature detection, image segmentation, object recognition, and 3D reconstruction.

 

Course prerequisite:


Announcements:


Lecturer: Prof. Jiaya JIA
Department of Computer Science and Engineering
Room 1018, Ho Sin-Hang Engineering Building
Tel: 3943-8396

Tutor: Mr. Xin TAO
Room 1026, Ho Sin-Hang Engineering Building
Email: xtao@cse.cuhk.edu.hk

Time and Venue:
Monday (14:30 - 16:15)  LSB LT4
Tuesday (13:30 - 14:15)  ERB 404

Reference Books:
- Computer Vision:  A Modern Approach, Forsyth & Ponce, Pearson, 2002
- Digital Image Processing, 2nd edition, Gonzalez and Woods
- Multiple View Geometry in Computer Vision, Hartley & Zisserman

 


Projects:

Projects

Deadline

Results

Assign. 1: spec, code 11:59pm, February 6th, 2017 Score, Quiz1
Assign. 2: spec, code, Szeliski's Textbook 11:59pm, February 22nd, 2017 Score
Assign. 3: spec, code v2 (March 8) 11:59pm, March 19th, 2017 Score
Assign. 4: spec, code_data_v3 11:59pm, April 9th, 2017  
Project: spec, code_data 11:59pm, May 10th, 2017  

 


Syllabus and lecture notes:

WEEK

CLASS  DATE

TOPICS

MATERIALS

1

9-10 Jan. Introduction

     

Camera and Calibration

2 16-17 Jan.

Image Processing

3 23-24 Jan.

Feature Detection and Matching

4 30-31 Jan. Happy Chinese New Year! (Holiday)

 
5 6-7 Feb.

Feature Detection and Matching (cont.)

  • We move to LSB LT4 (bigger room) for the Monday lecture since Feb. 6th.
6 13-14 Feb. Image Alignment and Mosaics

7 20-21 Feb. Motion estimation

8 27-28 Feb.

Professor on leave - Your Holiday Quiz Time

Quiz starts at 14:30 on 27 Feb and takes 1 hour top in the lecture room. Tutors will be there.

  • Lecture make-up will be on 24 April.
9 6-7 March Lighting and Photometric Stereo

10 13-14 March Detection and Simple Recognition

11 20-21 March Image Classification

 

12 27-28 March Convolutional Neural Networks for CV


  • slides: pptx

  • Stanford CS231n notes: Link
  • Tensorflow Playground: Link
  • 3D Visualization of CNN: Link
  • Deep Visualization: Link

 

13 3-4 April Convolutional Neural Networks (cont.)

4 April: Ching Ming Festival (Holiday)

14 10-11 April

Back Propagation

 
15 17-18 April 17 April: Easter (Holiday)

  • slides: pptx
  • Yolo-v2 (Real-time Object Detection): Link

  • Semantic Segmentation Demo: CRF as RNN, SegNet

  • Recent hot topics in deep learning:
  • Recurrent Neural Networks: Link
  • Reinforcement Learning: Link
  • Generative Models: Link
 

Detection and Segmentation

16 24 April Class make-up

Venue: ERB 405
Time: 14:30-17:15
 

 


Assessment:

FLTK Resources:

Installing FLTK
FLTK home page
FLTK Programming Manual
FLTK Demo for NT and its sources
FLTK Demo for Linux and its sources

Python Resources: