Seminar on Current Works in Computer Vision

Prof. Thomas Brox

Computer Vision is a very active research field with many practical applications, for instance in quality control, robotics, or driving assistance systems. The ultimate goal of Computer Vision is to imitate the great capabilies of the human visual system, allowing the computer not only to record images but also to interpret them. Research is still far from this goal, but significant progress has been made in recent years.
In this seminar we will take a detailed look at the most interesting recent works that have been published at the latest Computer Vision conferences CVPR 2012 and ECCV 2012. You will read a number of research papers published at these conferences. For each paper there will be one person, who performs a more detailed investigation of the research work and its background and gives a presentation. The presentation is followed by a discussion with all participants about the merits and limitations of the respective paper. You will learn to read and understand contemporary research papers, to give a good oral presentation, to ask questions, and to openly discuss a research problem.

(2 SWS)
Wednesday, 10-12am,
Room: HS 02-017, Building 052
Contact person: Robert Bensch

Beginning: Wednesday, Oct 24, 2012
Introduction and allocation of seminar topics.

ECTS Credits: 4

Recommended semester:

Requirements: any proseminar

Remarks: This course is offered to both Bachelor and Master students. Master students should give their presentation in English (to practise their English presentation skills), Bachelor students may present in German. The discussion will be in English but German questions and comments are welcome. Online evaluation from 28.1. to 8.2.

Please note: Submit your presentation outline to your advisor at least 2 weeks before your presentation and meet with your advisor. Submit your presentation slides to your advisor at least 1 week before your presentation and meet again.

All participants must read all papers and answer a few questions. The answers must be sent to the advisor before the paper is presented.

Slides of first session with instructions for a good presentation
Powerpoint template (optional)


Date   Topic Paper Questions  Presenting student   Slides   Advisor
05.12. Optical flow with occlusion modeling Unger et al. Questions Martin Goth Benjamin Drayer
12.12. Recognition with 3D object models Pepik et al. Questions Janosch Scharlipp Kun Liu
19.12. Synthetic movie for benchmarking Butler et al. Questions Brian Davis pdf Robert Bensch
09.01. Detector learning from videos Prest et al. Questions Anton Böhm Thorsten Schmidt
16.01. Registration of different object instances Kemelmacher et al. Questions Carl Faure pdf Benjamin Drayer
23.01. Clustering in part-based detection Divvala et al. Questions Manuel Bühler pdf Alexey Dosovitskiy
30.01. Object class detection Hoiem et al. Questions Dominik Gebhart Naveen Shankar Nagaraja
06.02. Object clustering and classification Hariharan et al. Questions Nikolaus Mayer pdf Philipp Fischer
13.02. Convolutional deep belief networks Huang et al. Questions Cem Uran pdf Ahmed Abdulkadir