MULTIMEDIA DATA MODELLING

1.  Knowledge and understanding: the objective of the course is for students to acquire knowledge that enables them to understand basic concepts about digital images and videos, as well as their processing for accurate analysis of low-level discriminative features useful for solving common tasks (face recognition; face detection; etc.)
2.  Ability to apply knowledge and understanding (applying knowledge and understanding): the student will acquire the necessary skills for processing and analyzing multimedia content. In this regard, part of the course will consist of laboratory lectures, with practical examples of programs written in Python language and use of ad-hoc libraries (mainly OpenCV)
3. Autonomy of judgment (making judgements): The student will be able to independently develop solutions in writing algorithmic solutions for the analysis of mutlimedia content.
4. Communication skills (communication skills): the student will acquire the main skills in the use of technical language in the general area of digital images and video and the basic concepts of computer vision.
5. Learning skills: the course aims, as an objective, to provide the student with the necessary theoretical and practical methodologies to be able to address and solve independently new problems that may arise during a work activity. To this end, various topics will be covered in class involving the student in the search for possible solutions to real problems.

• Basic knowledge of the Python language
• Basic concepts of Machine Learning

The first part of the course is about digital images:

• Introduction to digital images + Lab. Session
• Interpolation operations: replication, bilinear and bicubic + Lab. Session
• Space domain and Frequency domain + Lab. Session
• Fourier and DCT Transform  + Lab. Session
• The convolution and convolution theorem
• Lossy and lossless compression
• The JPEG standard
• Mathematical morphology applied to digital images.
• Mathematical morphology applied to gray-scale images + Lab. Session
• Image restoration and Noise models + Lab. Session
• Filters: arithmetic, geometric, harmonic and counter-harmonic mean + Lab. Session
• Median, minimum, maximum, midpoint, + Lab. Session
• Adaptive filters + Lab. Session
• Periodic noise. Noise removal in the frequency domain + Lab. Session
• Filtering in the spatial domain. Edge detector. Canny's algorithm + Lab. Session
• Steganography and Steganalysis + Lab. Session

The second part of the course is about digital video:

• Introduction to digital video and main definition: aspect ratio, resolution, video file format. 
• Lab. Session: OpenCV and Digital Video 
• Video formats (MPEG-1, MPEG-2, MPEG-4, H.264).

The third part of the course covers Low-Level Vision:

• Low-Level Vision: Filters and Features: Edges, Textures, Laplacian Pyramid, Corner Detection (Harris, ...), SIFT.
• Computer Vision applications: face detection and recognition, etc..

• Data modeling (features extracted from multimedia contents) and classification tasks
  

 Laboratory Session:

• The python language applied to digital image and video processing 
• Introduction to OpenCV and other image/video processing libraries 
• Implementation of Computer Vision algorithms (studied in the theoretical part)

Digital Image Processing, (3rd Edition) Rafael C. Gonzalez, Richard E. Woods, Ediz. Pearson, Prentice Hall

G. Bradski, A. Kaehler, “Learning OpenCV Computer Vision with the OpenCV Library” O'Reilly Media, 2008;

Mubarak Shah, "Fundamentals of Computer Vision" (pdf), 1997

Richard Szeliski, Computer Vision: Algorithms and Application, Springer 2010

1. What is a digital image?

a) A photograph taken with an analogue camera

b) A matrix of light intensity values

c) A compressed video

d) A hand-drawn image

2. What are the types of images?

a) Black and White, Gray and Blue

b) Gray and Green, Color, Pink and Blue

c) Black and White, Greyscale, Color

d) Yellow and Red, Green and Blue, Color

3. What does the RGB color space represent?

a) A series of abstract coordinates

b) A color representation based on red, yellow and blue

c) The gray line

d) A color representation based on red, green and blue

4. What is the spatial resolution of an image?

a) The number of colors present in the image

b) The number of information points (pixels) in the image

c) The physical size of the image

d) The speed at which the image is displayed

5. What are the three main types of frames in videos?

a) I, P, L

b) I, Q, B

c) I, P, B

d) I, X, Z