TeachingWs15TDCV

Chair for Computer Aided Medical Procedures & Augmented Reality
Lehrstuhl für Informatikanwendungen in der Medizin & Augmented Reality

Tracking and Detection in Computer Vision - Winter Term 2015/16

Administrative Info

Lecture by Dr. Slobodan Ilic
Exercises by: Christian Rupprecht, Mira Slavcheva

Type: Lecture Module IN2210
Programs: Informatics (Bachelor, Master)
Biomedical Computing (Master)
Robotics, Cognition, Intelligence(Master)
SWS: 2+4
ECTS: 7 Credits
Course Language: English

Time, Location & Requirements

Lectures: Monday 12:00 - 14:00 in 00.13.009A (MEDIEN)

Exercises: Thursday 12:00 - 14:00 in 00.13.009A (MEDIEN)

Consultation hours: Office of the responsible tutor (see names next to the exercise). To get an a slot for consultation send an e-mail to Christian Rupprecht.

Requirements:

  • The classes and exams are in English.
  • For the midterm exam, nothing but your pens and your calculators are allowed.
  • For the final exam, nothing but your pens, your calculators and one DIN A4 page (handwritten, front and back side) with notes are allowed.
  • Bonus points can be earned from the homework projects and the intermediate exam.
  • The final exam brings maximally 100 points. You need to have 50 points in order to pass it. Bonus points are irrelevant for the 50-points-hurdle and are added afterwards to your final exam score.
  • There is no repetition exam.

Contents

Overview

Computer Vision, as a relatively young research area, has emerged as a key discipline in computer science. This is not only evident by a growing and highly competitive research community with a high impact factor in computer science, but also by the emergence of numerous vision companies turning research ideas into a myriad of commercial applications. Besides well-known studies of 3D geometry and camera models, object tracking and detection in images and videos becomes one of the principal research directions of modern Computer Vision. The main objective of this course is to provide students with a gradual introduction to modern tracking, detection and recognition techniques developed in the last years. The course will provide in-depth knowledge of image features, their detection and description, matching techniques, key-point recognition, basic and advanced tracking algorithms based on image features and image intensities, basics of structure from motion and bundle adjustment, synchronous localization and mapping algorithms as well as object detection using classification approaches. In the end, the students will have a thorough description of the most important tracking and detection techniques. They should be able to understand and implement those solutions and apply them in reasonably complex problems. The concepts described in this course will be accompanied with brief explanations of the necessary mathematical tools. The participants of this course will be given an important basis to follow the vast and growing Computer Vision literature and use the acquired knowledge to solve new practical problems.

Who should attend this course?

Bachelor students in their last years and all Master students that are interested in learning about the modern and well-established concepts and algorithms related to tracking, detection and recognition in Computer Vision, or desire to use those techniques in their research.

Prerequisites: Most of the knowledge required should be part of the normal background in Computer Science, undergraduate/graduate Mathematics and Geometry.

Announcements

* The final exam inspection (Klausureinsicht) will be on Thursday 25th of February at 3-4:30pm in 03.13.010.
* The overview of exercise mid- and end-term scores can be found here.
* The boni (exercise and midterm) are applied if the final exam was passed (>= 50pts).
* The grades are distributed with 5pt increments. E.g. 1.0: >= 95pts, 1.3: >= 90pts, 1.7: >= 85pts etc.

Exercises

Exercise FAQ

The exercises will be done in MATLAB and students are expected to do them on their own computer/laptop. In case you do not have your own laptop please let us know by sending an e-mail to: tdcv.tum@gmail.com.

You can also find a short MATLAB tutorial here.

Registration

You are required to register via TUMOnline (link)!

Homework

Students will be given one week to solve the problems. The exact due dates for each homework are announced next to the exercises on this website and will be mentioned during the exercise sessions. Students have to make groups of TWO students and send the names to tdcv.tum@gmail.com. After that, those who didn't create groups will be associated to groups based on a random selection and will receive a notification about this. If you are looking for a (TDCV) partner tell us and we will assign you one!

During the checking of the exercises each group will have at 15 min to be questioned to answer the questions. All members of the group will be questioned about the algorithms and their implementation, so only those who can prove that they understood both theory and practice will get the respective points. Thus, having a working exercise doesn't mean that everyone in the group is going to get full points. Please note that only exercises which work might get the maximal number of points. There will be no "debugging-correcting"-sessions for your solutions during the exercise reviews! If you have questions, ask them beforehand by either mail or in person.

All homework should be submitted through e-mail at least one hour before the exercise session with the following format and information.

  1. To: tdcv.tum@gmail.com
  2. Subject: [TDCV15] Exer<exercise number> - <last names of all members>
    For example: [TDCV15] Exer1 - Huang, Tan
  3. Attachment: The whole source code for this homework.
  4. Body: This should include the (a) full names and Matrikel-Nr. of each member; and, (b) reference to any code used in the homework that is not specified on the exercise sheet or by the tutors.

Textbooks

For some basic concepts the textbook "Computer Vision: a modern approach" by David Forsyth and Jean Ponce can be used. Also, most topics can be found in a new book of Rick Szeliski from Microsoft Research, available also on-line. However, since the majority of the topics is state-of-the-art, students will be referred to the relevant publications which will be provided during the course and can be regarded as additional lecture notes.

Lecture Schedule and Content

Date Topic Conducted by Material/Literature
Mo 19.10.2015 12:00 - 14:00 Introduction Dr. S. Ilic Introduction
Mo 26.10.2015 12:00 - 14:00 Convolution, Non-linear filtering, Gaussian Filtering, Image Derivatives, Edge Detection Mira Slavcheva Filtering and Edge Detection
Mo 02.11.2015 12:00 - 14:00 Local invariant feature detectors: Harris corner, Harris Laplace/Affine, Hessian, Hessian-Laplace/Affine, FAST Dr. S. Ilic Feature Detection
Related papers:
1. Feature detectors
2. Harris-Laplace
Mo 09.11.2015 12:00 - 14:00 Feature descriptors: Difference of Gaussians and SIFT, Integral images and SURF, Histogram of Oriented Gradients Dr. S. Ilic Feature descriptors
Related papers:
1. SIFT
2. SURF
3. HOG
Mo 16.11.2015 12:00 - 14:00 Keypoint recognition: Randomized and regression trees, FERNS, Hough Forests Dr. S. Ilic Keypoint recognition
Related papers:
1. Randomized Trees
2. FERNS
3. Hough Forest
4. Decision Trees
Mo 23.11.2015 12:00 - 14:00 Camera models, Model based tracking, Pose estimation from 2D-3D coresspondencies (DLT, P-n-P), Rotation parametrization Dr. S. Ilic Camera models and pose estimation
Related paper:
Monocular Model-Based 3D Tracking of Rigid Objects: A Survey
Mo 30.11.2015 MIDTERM EXAM Dr. S. Ilic Mid-Term Example
Mo 07.12.2015 12:00 - 14:00 Non-linear optimization, Robust estimators, RANSAC Dr. S. Ilic Non-linear Optimization, RANSAC
Related paper:
Monocular Model-Based 3D Tracking of Rigid Objects: A Survey
Practical Parameterization of Rotations . F. S. Grassia
Mo 14.12.2015 12:00 - 14:00 Haar features, Ada-boost, Viola-Jones Face Detection Dr. S. Ilic Face Detection
Related papers:
1. Viola-Jones
Mo 21.12.2015 12:00 - 14:00 Template tracking methods: Lucas-Kanade, Compositional Alg., Inverse Compositional, ESM, Linear Predictor Dr. S. Ilic Template tracking methods: Lucas-Kanade,Compositional Alg., Inverse Compositional, ESM, Learning a linear predictor, ASM and AAA
Related paper:
1. Lucas-Kanade 20-Years On - A Unifying Approach
2. Hyperplane approximation for template tracking
3. ASM + AAM
Mo 11.01.2016 12:00 - 14:00 Mean-shift tracking Dr. S. Ilic Mean-shift Tracking
Related paper:
1. Mean Shift: A Robust Approach Toward Feature Space Analysis
2. Kernel-Based Object Tracking
3. Mean-shift Blob Tracking through Scale Space
Mo 18.01.2016 12:00 - 14:00 Template matching approaches Dr. S. Ilic Template Matching
Related paper:
1. DOT
2. LineMod
Mo 25.01.2016 12:00 - 14:00 Chameleon Tracker Dr. S. Ilic Chameleon Tracker
Related paper:
1. Multi-Forest Tracker: A Chameleon in Tracking
Mo 01.02.2016 12:00 - 14:00 Summary Dr. S. Ilic Canceled

Exercise Schedule

Date Topic Conducted by Material/Literature Sample Output
Do 22.10.2015 12:00 - 14:00 Matlab Introduction/Convolution, Gaussian Filtering, Image Derivatives Mira & Christian Exercise01
Lena.gif
lenanow.jpg
Sample Output 1
Sample Output 2
Sample Output 3
Do 29.10.2015 12:00 - 14:00 Median and Bilateral Filtering Mira Exercise02 Sample Output 1
Sample Output 2
Do 05.11.2015 12:00 - 14:00 Harris, Harris-Laplace Leslie & Paul Exercise03
house.tif
checkerboard_tunnel.png
test.pgm
sample2.jpg
Sample Output 1
Sample Output 2
Sample Output 3
Sample Output 4
Do 12.11.2015 12:00 - 14:00 Textured and texture-less object detection with SIFT and HOG David Exercise04
Files
 
Do 19.11.2015 12:00 - 14:00 Random Forests Christian Exercise05 (updated)
Files
Slides
Sample Output 1
Do 26.11.2015 12:00 - 14:00 Sift, Dlt, Ransac, Image Stiching Tolga Exercise06
Files
Sample Output 1
Sample Output 2
Do 10.12.2015 12:00 - 14:00 Non-linear optimization (due: 17.12.) Paul Exercise07 (updated)
Sequence
Slides and sample output
 
Do 17.12.2015 12:00 - 14:00 Ada-boost classification (due: 7.1.) Nikolas Exercise08
Data
Sample Outputs
Do 07.01.2016 12:00 - 14:00 Template Tracking with Linear Predictor (due: 14.1.) Tolga Exercise09
Data
For more details, see section 3 of:
Online Learning of Linear Predictors for Real-Time Tracking
Do 14.01.2016 12:00 - 14:00 Mean-shift Tracking (due: 21.1.) David Exercise10
Data
 
Do 21.01.2016 12:00 - 14:00 Template matching (due: 28.1.) Wadim Exercise11 Sample Output
Do 28.01.2016 12:00 - 14:00 Template matching corrections - - -

If you have questions about a specific exercise, please direct them to the responsible tutor.

TeachingForm
Title: Tracking and Detection in Computer Vision
Professor: Dr. Slobodan Ilic
Tutors: Christian Rupprecht, Mira Slavcheva
Type: Lecture
Information: 2+4 SWS, 7 ECTS,
Term: 2015WiSe
Abstract:  


Edit | Attach | Refresh | Diffs | More | Revision r1.86 - 18 Jan 2017 - 15:15 - TolgaBirdal

Lehrstuhl für Computer Aided Medical Procedures & Augmented Reality    rss.gif