Revision f8bcfb19bc0ea24eadb41b012112ba943c84ca15 (click the page title to view the current version)

Front Page

Changes from f8bcfb19bc0ea24eadb41b012112ba943c84ca15 to current

---
title: AIS2204 Maskinsyn
categories: Module
---

+ See [Lecture Notes](Overview) for learning materials
+ See [Blackboard](https://ntnu.blackboard.com/) for announcements and 
  discussion boards.

# Practical Information

1.  There are no compulsory exercises.  It is your responsibility to
    the exercises you need to do to understand the subject and gain
    experience.
2.  Feedback is provided *in class* upon demonstration of own work
    and solutions.
3.  The module emphasises the relation between theoretical and practical
    understanding.
4.  I try to give you *freedom to learn*.  Please use that freedom to
    learn what you need to know.

## How to work with the module

1.  Read the theory.
2.  Do practical exercises to *test your theoretical understanding*
    - Do the exercises that you need to do to make sense of the theory.
3.  Evaluate your own solutions and reflect upon
    - what have you learnt from the exercise?
    - what do you yet not know?
4.  Don't do a lot of exercises quickly.
    It is better to make sure that you comprehend a few exercises fully,
    and can justify and validate your own reasoning.
5.  Ask Questions.  
    I will generally not repeat material unsolicited, 
    but I am very happy to discuss any question you may have.
4.  Keep a diary.
    Make sure you can refer back to previous ideas and reuse previous solutions.

## The practical exercises

Both practical and theoretical exercises are given.
Most of the exercises are designed in the hope that you can
complete them in a session or two, but also move on even
if you do not complete them.

Observe that the exam gives you a lot of freedom to emphasise
what you find interesting and useful.  Thus you will be rewarded
for solving related variants and for tying different exercises 
together in more complete systems.

## How does the exam work

1.  Oral Exam.
2.  You get seven minutes to demonstrate the highlights of your
    understanding of the subject.
    *Make a case for the grade you think you deserve.*
3.  The examiner will use the rest of the time for questions to clarify
    and to demonstrate expected breadth and depth.
4.  Note that there are both theoretical and practical learning outcomes,
    and the module emphasises the relation between these two.

+ Capacity: 30 candidates
+ [Assessment Guide]().
  This will be reviewed in the reference group and only made
  final at the end of the teaching term.
 

# Syllabus

+ The syllabus is defined by the lectures and exercises.
+ Two textbooks will be used, both available electronically for free:
    - [Ma (2005): An Invitation to 3-D Vision: From Images to Geometric Models](https://doi.org/10.1007/978-0-387-21779-6)
        - Chapters 1-5 and 11
        - This book gives a very good presentation of the fundamental theory, but it is unfortunately
          outdated when it comes to current applications.
    - [Szeliski (2022): Computer Vision: Algorithms and Applications, 2nd ed.](https://szeliski.org/Book/)
        - Chapters to be announced.
        - This book is very up to date, but it does not always describe the algorithms in detail.
        - Chapter 5.1-5.4. 
        - Cursory Chapters 5-6.
        - Chapter 7.1-7.4.
          (This partly overlaps with chapters from Ma (2005),
          but the coverage of feature descriptors is particularly
          important.)
        - This book is up to date and covers feature descriptors and
          machine learning.  
          Unfortunately it does not always describe the algorithms
          in detail.
- Additional reading:
    + [Alexey Spizhevoy (2018) *OpenCV 3 Computer Vision with Python Cookbook*](https://bibsys-almaprimo.hosted.exlibrisgroup.com/primo-explore/fulldisplay?docid=TN_cdi_safari_books_v2_9781788474443&context=PC&vid=NTNU_UB&lang=no_NO&search_scope=default_scope&adaptor=primo_central_multiple_fe&tab=default_tab&query=any,contains,Alexey%20Spizhevoy&offset=0)
      Available electronically from the university library; see [Oria](https://oria.no/).