Number of hours
- Lectures 18.0
- Projects -
- Tutorials -
- Internship -
- Laboratory works -
- Written tests -
ECTS
ECTS 3.0
Goal(s)
This course presents the evolution of learning methods for recognition over the past two decades, starting with techniques using hand-crafted features to modern feature learning algorithms.
Responsible(s)
Karteek ALAHARI
Content(s)
Methods for several problems in recognition, such as semantic segmentation, object detection, action detection, etc., will be discussed in detail.
The topics discussed include:
- Features for recognition
- Matching algorithms
- Clustering methods
- Generative and discriminative models
- Bag-of-features for retrieval and classification
- Graphcuts
- Deep learning
- Action recognition
Test
Quiz organized during the lecture sessions and a final exam.
N1=50%EXAM1+50%QUIZ/CC
N2=50%EXAM2+50%QUIZ/CC
The exam is given in english only 
Calendar
The course exists in the following branches:
- Curriculum - Master 2 in Computer Science - Semester 9 (this course is given in english only
) - Curriculum - Master 2 in Applied Mathematics - Semester 9 (this course is given in english only )
- Curriculum - Master 2 in Computer Science - Semester 9 (this course is given in english only )
Additional Information
Course ID : WMM9MO31
Course language(s):
You can find this course among all other courses.
Bibliography
Papers referenced on the course website.