| Course title: | Reading group in Pattern Recognition and Computer Vision |
| CTU code: | XP33RCV |
| Course link: | https://fel.cvut.cz/en/education/bk/predmety/60/19/p6019906.html |
| Semester: | Winter (formally, but the course continues in the summer semester). |
| Teacher: | Giorgos Tolias |
| Guarantor: | Ondrej Chum |
Annotation
The course deals with fundamental results in computer vision and pattern recognition. It targets the detailed study of principles which substantially influence the development in the field. The course is performed in the form of a reading group. Each time, a person in charge presents a paper and the reading group participants join in with questions, comments and discussion about the paper. Each student is expected to present one paper and actively participate in reading groups of other presenters.
Schedule
Reading groups are organized irregularly, several times a year. The reserved time slots are those for CMP seminars, on Tuesday or Thursday 11:00-12:30. Each reading group is announced separately here.
Meetings are included in the following calendar: cmp.felk.cvut.cz/events/icalc.phtml.
Currently all meetings are online. Links will be included in the announcements.
Reading group instructions
Selecting a paper
Students can select a paper from the preselected list or make their own suggestion. The suggestion should be consulted with the supervisor and has to be approved by the teachers. Typical sources are the following conferences and journals: CVPR, ICCV, ECCV, PAMI, IJCV, NeurIPS, ICLR, and JMLR.
Reading group participants
The meetings are open to anyone, disregarding their background. It is assumed that everyone attending the reading group has, at least briefly, read the paper – not necessarily understanding everything. During the presentation we aim to have a fruitful discussion, a critical analysis of the paper, as well as brainstorming for creative extensions.
Preparation of the presenter
- Read and try to understand the paper in detail.
- For the parts that you find difficult, at least be ready to explain what is not clear and why.
- Understand and get prepared to present the addressed problem, and the findings of the authors.
- Think about how this work affects what we knew so far about the problem.
- Identify aspects which are not handled well or any unrealistic assumptions.
- Prepare a slide presentation; feel free to use any available material from the authors.
- Explain necessary background knowledge, which is assumed by the authors but probably not known by everyone.
- Check for recent progress/extensions (look into papers citing this work).
- Have a look at previous presentations: videos can be found in the main page.