This page is located in archive.


Student and course evaluation

Course outline

date week topic slides
23.09.2019 1 Overview and lecture outline, Regression as ML/MAP estimate vir_outline.pdf
30.09.2019 2 Classification as ML/MAP estimate learning_for_vision_i.pdf
07.10.2019 3 Neural networks + Convolutional layer + backpropagation learning_for_vision_ii_neural_nets.pdf learning_iii_convnets.pdf
14.10.2019 4 Test T1 + Training I (SGD, momentum, convergence rate)

21.10.2019 5 Training II (Adagrad, RMSProp, AdamOptimizer, diminishing/exploding gradient, oscilation)
28.10.2019 6 Holidays
04.11.2019 7 Elementary layers: (Activation functions, Batch-Instance Norm, MaxPooling, Losses) + backpropagation learning_for_vision_v_layers.pdf
11.11.2019 8 Architectures of deep neural networks I: detection (Yolo), segmentation (DeepLab), classification (ResNet)

18.11.2019 9 Test T2 + Architectures of deep neural networks II: pose regression, action recognition (recurrent nets), LIFT (spatial transformer nets), Style Transfer Networks
25.11.2019 10 Reinforcement Learning I (DQN, GAE+TD(lambda), Inverse RL, REINFORCE, DDPG, Actor-Critic, applications) reinforcement_learning.pdf
02.12.2019 11 Reinforcement Learning II (REINFORCE, DDPG, Actor-Critic, applications)
09.12.2019 12 Generative Adversarial Networks (guest lecture: David Coufal, UTIA) gans_coufal_vir_2019
16.12.2019 13 Exam Test ET see BRUTE for results
06.01.2020 15 Presentation of selected semestral works

Course organization and credit conditions

The lectures takes place in KN:E-301 on Monday 14:30-16:00. The labs will take place on Monday, Tuesday and Wednesday, see schedule for details. The first 7 weeks will be focused on lectures, tests and homework, the remaining 7 weeks are devoted to the semestral work. Lecture in the 13th week is devoted to exam test. The last lecture will be devoted will consist of the presentations of (selected) semestral works. The points are divided into homework (30), semestral work (30), tests & exam (40) .

Minimum credit requirements:

  • obtain at least 50 points in total.
  • achieve at least one point from each homework
  • achieve at least one point from exam test

The final grade will be determined by the total number of points according to the following table

No of points Exam assessment
0-49 F
50- 59 E
60-69 D
70-79 C
80-89 B
90-100 A


There will be three automatically evaluated homework in total. Maximum number of points from homework is 30. All homework will be assigned during the labs, see labs schedule for the assignment dates. The submission of each homework has strict deadline. Initial homework HW0 will be for 6 points, following two homework are for 12 points each.

Semestral work

Each semestral work will be solved by the team of three students. Maximum number of points for the semestral work is 30. There will be a limited list of topics available by internal and external supervisors, however students are encouraged to come up with their own SW topic. If more than one team is interested in supervisor's topics, the assignment will be based on the average number of points achieved by all team members from T1+HW1.

supervisors topics:

  • topics/supervisors assigned during the labs in the 7th week (it is desirable, that the team is capable to consult SW assigned by Ota on Monday's labs, Patrik on Tuesday's lab and Teymur on Wednesday's lab).
  • limited capacity ⇒ assignment based on the average number of points of all team members achieved from T1 and HW1.
  • progress will be monitored based on the individual agreement between the team and the supervisor.

own topics:

  • unlimited capacity ⇒ available for all students
  • supervisor is the lab tutor responsible for particular labs as follows: Ota for Monday's labs, Patrik for Tuesday's labs and Teymur for Wednesday's labs
  • choice of the topic is part of the evaluation process
  • milestone M1 (9th week): topic approved by supervisor
  • milestone M2 (12th week): preliminary results demonstrated.

The presentation takes place in the 14 week. The work will be presented to all students on the poster session during the Monday lecture. In addition to this, there will be 10 minute long oral presentations, which will take place during the labs. Evaluation is based on the supervisor's, lecturer's and student's voting as follows:

  • 15 from supervisors
  • 10 from lecturers
  • 5 from other students


There will be three tests in total, which will take place during Monday Lectures, see schedule for planned weeks. Maximum number of points from tests is 30: first two tests (T1 and T2) each for 10 points, exam test (ET) for 20 points. Competencies required for passing the test will be summarized in the end of each lecture.


http://cmp.felk.cvut.cz/~zimmerk Karel Zimmermann is the main lecturer of ViR. He is currently associate professor at the Czech Technical University in Prague. He received his PhD degree in cybernetics in 2008. He worked as postdoctoral researcher with the Katholieke Universiteit Leuven (2008-2009) in the group of prof Luc van Gool. His current H-index is 13 (google-scholar) and he serves as a reviewer for major journals such as TPAMI or IJCV and conferences such as CVPR, ICCV, IROS. He received the best lecturer award in 2018, the best reviewer award at CVPR 2011 and the best PhD work award in 2008. His journal paper has been selected among 14 best research works representing Czech Technical University in the government evaluation process (RIV). Since 2010 he has been chair of Antonin Svoboda Award (http://svobodovacena.cz). He was also with the Technological Education Institute of Crete (2001), with the Technical University of Delft (2002), with the University of Surrey (2006). His current research interests include learnable methods for robotics.

http://cmp.felk.cvut.cz/~azayetey Teymur Azayev is the head of ViR's labs. He is the second year PhD student at the Department of Cybernetics. His current interests include dynamic robot locomotion using deep learning.

http://cmp.felk.cvut.cz/~vacekpa2 Patrik Vacek will be assisting in VIR labs. He is the first year PhD student at the Department of Cybernetics. His main research interests include deep learning for self-driving cars.

http://cmp.felk.cvut.cz/~jasekota Otakar Jašek will be assisting in VIR labs. He is the second year PhD student at the Department of Cybernetics. His main research interests are about processing point clouds by the means of deep learning.


We want students to work individually, therefore any plagiarism in codes, homework or reports will be mercilessly punished ;-). We strongly urge each student to read what is/isnot a plagiarism - we believe that many students will be surprised. In any case, it is not permitted to use the work of your colleagues or predecessors. Each student is responsible for ensuring that his work does not get into the hands of other colleagues. In the case of multiple submission of the same work, all involved students will be penalized, including those who gave the work available to others

courses/b3b33vir/start.txt · Last modified: 2020/01/14 12:17 by zimmerk