Tracking and Detection in Computer Vision
Module IN2210
This module handbook serves to describe contents, learning outcome, methods and examination type as well as linking to current dates for courses and module examination in the respective sections.
Module version of SS 2015 (current)
There are historic module descriptions of this module. A module description is valid until replaced by a newer one.
Whether the module’s courses are offered during a specific semester is listed in the section Courses, Learning and Teaching Methods and Literature below.
| available module versions | |
|---|---|
| SS 2015 | WS 2011/2 |
Basic Information
IN2210 is a semester module in English language at Bachelor’s level and Master’s level which is offered in winter semester.
This Module is included in the following catalogues within the study programs in physics.
- Catalogue of non-physics elective courses
| Total workload | Contact hours | Credits (ECTS) |
|---|---|---|
| 210 h | 90 h | 7 CP |
Content, Learning Outcome and Preconditions
Content
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 probabilistic and machine learning methods for tracking and object detection. Note that every year the course content is refreshed with new the most promising and potentially the most influential works in the field.
The following topics will be handled
- Introduction (overview of the course)
- Convolution and filtering:
-- Basic of image formation
-- Convolution and correlation
-- Non-linear filtering
-- Gaussian Filtering
-- Image Derivatives
-- Edge Detection
- Local invariant feature detectors:
-- Harris corner
-- Harris Laplace/Affine
-- Hessian, Hessian-Laplace/Affine
-- FAST
- Feature descriptors:
-- Difference of Gaussians and SIFT
-- Integral images and SURF
-- Histogram of Oriented Gradients (HOG)
- Keypoint recognition:
-- Randomized trees
-- FERNS
-- Keypoint signatures
- Face detection
-- Haar features
-- Ada-boost
-- Viola-Jones Face Detection
- Camera models and projections
-- Model based tracking
-- Pose estimation from 2D-3D coresspondencies (DLT, P-n-P)
-- Rotation parametrization
- Non-linear optimisation
-- Robust estimators
-- RANSAC
- Template tracking methods:
-- Lucas-Kanade,
-- Compositional Alg.
-- Inverse Compositional
-- ESM
-- Linear Predictor
- Mean-shift tracking
-- mean-shift for pdf estimation
-- mean-shift for segmentation
-- mean-shift for object tracking
-- multi-scale
- Template matching approaches
-- basic correlation methods (SAD, NCC etc.)
-- DOT(Dominant Orientation Template)
-- LineMod (LINEarizing the memory multiMODal template matching)
- Kalman and particle filtering
-- basics Kalman filer
-- basics Particle filer
-- applications to visual tracking
-- applications to camera tracking
- Tracking with Dictionary Learning
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 probabilistic and machine learning methods for tracking and object detection. Note that every year the course content is refreshed with new the most promising and potentially the most influential works in the field.
The following topics will be handled
- Introduction (overview of the course)
- Convolution and filtering:
-- Basic of image formation
-- Convolution and correlation
-- Non-linear filtering
-- Gaussian Filtering
-- Image Derivatives
-- Edge Detection
- Local invariant feature detectors:
-- Harris corner
-- Harris Laplace/Affine
-- Hessian, Hessian-Laplace/Affine
-- FAST
- Feature descriptors:
-- Difference of Gaussians and SIFT
-- Integral images and SURF
-- Histogram of Oriented Gradients (HOG)
- Keypoint recognition:
-- Randomized trees
-- FERNS
-- Keypoint signatures
- Face detection
-- Haar features
-- Ada-boost
-- Viola-Jones Face Detection
- Camera models and projections
-- Model based tracking
-- Pose estimation from 2D-3D coresspondencies (DLT, P-n-P)
-- Rotation parametrization
- Non-linear optimisation
-- Robust estimators
-- RANSAC
- Template tracking methods:
-- Lucas-Kanade,
-- Compositional Alg.
-- Inverse Compositional
-- ESM
-- Linear Predictor
- Mean-shift tracking
-- mean-shift for pdf estimation
-- mean-shift for segmentation
-- mean-shift for object tracking
-- multi-scale
- Template matching approaches
-- basic correlation methods (SAD, NCC etc.)
-- DOT(Dominant Orientation Template)
-- LineMod (LINEarizing the memory multiMODal template matching)
- Kalman and particle filtering
-- basics Kalman filer
-- basics Particle filer
-- applications to visual tracking
-- applications to camera tracking
- Tracking with Dictionary Learning
Learning Outcome
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.
Preconditions
Most of the knowledge required should be part of the normal background in Computer Science, undergraduate/graduate Mathematics and Geometry.
Courses, Learning and Teaching Methods and Literature
Courses and Schedule
| Type | SWS | Title | Lecturer(s) | Dates | Links |
|---|---|---|---|---|---|
| VI | 6 | Tracking and Detection in Computer Vision (IN2210) | Bui, L. Ilic, S. |
Mon, 14:00–16:00, MI 00.13.009A Thu, 14:00–16:00, MI 01.05.012 Thu, 14:00–16:00, MI 03.11.018 Thu, 14:00–16:00, MI 00.13.009A Thu, 14:00–16:00, MI 03.09.012 and singular or moved dates |
eLearning documents |
Learning and Teaching Methods
Lecture, combined with experimental programming assessment targeting practical implementations of the methods explained at the lectures. The practical programming assessments are first explained at the exercises and then given in the form of home works for students to do them in groups.
Media
slides, blackboard, programming experiments
Literature
- lecture slides
- accompanied scientific papers and book excerpts
- Computer Vision: a modern approach" by David Forsyth and Jean Ponce
- Computer Vision: Algorithms and Applications, Rick Szeliski
- accompanied scientific papers and book excerpts
- Computer Vision: a modern approach" by David Forsyth and Jean Ponce
- Computer Vision: Algorithms and Applications, Rick Szeliski
Module Exam
Description of exams and course work
Type of Assessment: written exam (105 minutes)
Mid-term exam is planed after 6 lectures and it is written exam. The final exam takes place at the end of the lectures and it also takes the form of written test. Questions allow to assess acquaintance with concepts thought at the lectures and exercises, small calculation tasks and pseudo codes of the algorithmic solutions.
Programming tasks are not part of the examinations.
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.
Mid-term exam is planed after 6 lectures and it is written exam. The final exam takes place at the end of the lectures and it also takes the form of written test. Questions allow to assess acquaintance with concepts thought at the lectures and exercises, small calculation tasks and pseudo codes of the algorithmic solutions.
Programming tasks are not part of the examinations.
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.