Notice
This page is located in a preparation section till 23.09.2019.

Links

Course organization and credit conditions

The course takes place in KN:E-301 on Monday 14:30-17:30, see schedule for details. We do not distinguish the slot for lecture and slot for labs. The first 7 weeks are solely devoted to lectures, tests and homework, the remaining 7 weeks are devoted to the semestral work. The last lecture will consist of (i) the presentation of semestral work and (ii) the exam test. You can get 50 points from tests and homework and 50 points from the semestral work. Minimum credit requirement is to obtain at least 50 points in total. The final grade will be determined by the total number of points.

Semestral work

Each semestral work will be solved by the team of three students. Available topics will be announced in the 5th week. The presentation will take place in the last week (7.1.2019). Maximum number of points you can get from the semestral work is 50. Evaluation is based on the supervisor's, lecturer's and student's voting as follows:

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

Course outline

date week due to topic slides
01.10.2018 1 T0 Outline, organization of the course. vir_outline.pdf
Linear regression and classification + priors learning_for_vision_i.pdf
08.10.2018 2 Neural nets + backprop.
Assignment of HW0
learning_for_vision_ii_neural_nets.pdf
Convolutional network + backprop learning_iii_convnets.pdf
15.10.2018 3 T1, HW0 Learning (gradients, adaptive methods) learning_for_vision_iv_training.pdf
Layers (activation functions, losses, regularization layers) learning_for_vision_v_layers.pdf
22.10.2018 4 Frameworks for learning
Assignment of HW1
DL software overview
Pytorch tutorial
29.10.2018 5 T2 Guest lecture: Tomas Krajnik - Features features_navigation_slides
Guest lecture: Tomas Krajnik - Navigation
05.11.2018 6 HW1 Architectures for classification, regression, segmentation, object detection and feature matching in images learning_for_vision_v_architectures.pdf
Camera with radial distortion (calibration as optimization problem)
Assignment of HW2
camera.pdf
12.11.2018 7 Depth sensors (stereo, kinect, time-of-flight) exteroceptive_sensors.pdf
Mapping (basic SLAM, recurrent nets, learnable SLAM, stereo and monodepth) mapping_i.pdf, learnable_slam_recurrent_nets.pdf
Motion control from images reinforcement_learning.pdf
Domain transfer (GANs, time-constrastive nets, style transfer nets, inception score). Assignment of SW domain_transfer.pdf
19.11.2018 8 — supervisor consultations —
26.11.2018 9
03.12.2018 10
10.12.2018 11
17.12.2018 12
07.01.2019 15 SW, T4 Semestral work presentations
Exam test

Lecturers

http://cmp.felk.cvut.cz/~zimmerk Karel Zimmermann is the main lecturer of ViR. He is recently the associate professor at the Czech Technical University in Prague. He received his PhD degree in cybernetics from the Czech Technical University in Prague, Czech Republic, in 2008. He worked as the postdoctoral researcher with the Katholieke Universiteit Leuven (2008-2009). He serves as a reviewer for major journals such as TPAMI or IJCV and conferences such as CVPR and ICCV. He received the best reviewer award at CVPR 2011 and the CSKI prize for the best PhD work in 2008. 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 computer vision and robotics.


http://cmp.felk.cvut.cz/~petrito1 Tomas Petricek is the head of ViR's labs. He is the postdoctoral researcher in the Toyota Research Lab at the Czech Technical University in Prague. His main research interests include learnable mapping from RGB(D) sensors.








http://cmp.felk.cvut.cz/~azayetey Teymur Azayev will be assisting in VIR labs. He is a PhD. student at the Faculty of Cybernetics. His current interests include dynamic robot locomotion using Deep Learning.








Plagiarism

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: 2019/04/23 13:44 by zimmerk