The field of Computer Vision strives to develop algorithms for understanding, interpreting and reconstructing information of real world scenes from image and video data. On the other hand, the focus of Computer Graphics has been on the image syntehsis side, i.e., the development of algorithms to build and edit static and dynamic vrtual worlds and to display them in photorealistic or stylized ways.
In recent years, it could be observed that the fields have more and more converged since both sides started to investigate research questions that are of relevance for either side. As an example, Computer Graphics researchers have started to investigate algorithms to reconstruct detailed models of static and dynamic scenes from image data, such that more believable virtual renderings of real world scenes could be achieved. In a similar manner, Compter Vision research has been able to benefit from the experience of Computer Graphics research to efficiently and effectlively model image fomration and light transport, which also simplifies scene analysis and reconstruction tasks. Recently, the ever increasing amount of image and video data has further strengthened the links between the two fields, as both communities start to investigate new ways to analyze, structure and immersively display these data.
This seminar covers advanced research topics that lie on the boundary betweem the fields of Computer Vision and Computer Graphics. This includes classic and recent research results that were published at the top tier journals and conferences in either field. More specifically, the seminar will cover research papers from the following problem fields: Motion estimation and tracking, multi-view geometry and reconstruction, computational photography and videography, reonstruction of static and dynamic 3D scenes, advanced image and video processing, etc.
The target audience are graduate students in computer science or related fields. Basic knowledge in at least one of the fields of 3D geometry, computer vision, image processing, and computer graphics is required. Every participant will give a talk on a chosen scientific topic. Afterwards, the topic will be discussed within the seminar group. This is a great opportunity for students to improve their research and presentation skills and to learn about the latest developments in computer vision and computer graphics. The seminar language is English.
Every participant will perform a detailed reading on one topic and present the main ideas in class in an approximately 45 min presentation.
Working on one topic usually requires reading and understanding of the papers to be presented as well as acquiring background information necessary to understand the topic. For background knowledge usually some of the papers/books referenced in the paper will have to be consulted.
The presentation should cover the main ideas of the chosen topic. Usually, this means that content has to be selected from the paper; it will often not be possible to present all results of the two articles. Try to focus on the topic instead. Find a common thread linking the two papers and build your presentation on that. If you use formulas in your presentation make sure that all symbols are introduced properly. Similarly, make sure that figures are labeled correctly and that all terminology that is new to the topic is introduced appropriately. You can of course assume that your audience is familiar with the topics that have already been presented in class. It is very important for a good presentation to find a good balance between overview and detail. Most importantly, the general theme of your topic should become clear from your presentation. Your goal is that the other participants have a general knowledge about your topic after your presentation. You can pick one or two things that you find interesting and present those in more detail. Finally, prepare your presentation on time. Our feedback will help you to improve it. Also plan to spend some time on practicing your talk.
Important: We expect you to lead the discussion after your presentation. This requires that you prepare some questions, remarks or provocative statements for the audience. Try to challenge them. The discussion is also a good place to collect information and thoughts for the paper section where you present your own ideas. Keep this in mind when preparing your questions.
In addition to in class presentations, we require a written report on the topic. This report should summarize the main ideas of the paper and discuss limitations and drawbacks of the work. Based on this analysis, participants should develop and sketch an idea of their own of how to address one specific shortcoming. The report should consist of 6-8 pages for the material covering the presented papers and about 2-3 pages for the improvement proposed by the student.
The paper uses a template for one of the two major computer vision conferences, the IEEE Conference on Computer Vision and Pattern Recognition. Please use the review version of this template. It makes it easier to refer to certain parts of your text if need arises. For your convenience we provide the template links here:
.tar.gz (Linux) .zip (Windows)
Your paper should be emailed to us by August 11 2011. The format is .pdf. Please keep that in mind when using the Word template.
We will grade your paper based on a number of criteria. Again, as in the presentation, focus on getting the "big picture" right. Present your topic in context. Try to link the two articles, present a coherent text, i.e. a structure of simply
Paper 1
Paper 2
is usually insufficient. Also, try to avoid copying the structure of the original papers as far as this is possible. Equations can of course be reproduced, but do not stick too closely to the original text. Use your imagination to find a way for presenting the topic in your own words. The articles have been chosen such that they provide different views on a common theme. Try to exploit that link. Also keep in mind to structure your paper well in a formal sense. An abstract, summarizing the paper provides a short overview. The introduction expands on that. Give an overview of your paper's structure. After that present your topic and details of interest to you. Then expand on the topic, i.e. add your own ideas. Finally, a conclusion summarizes the paper, this is also the place where you can add your own opinions about the topic.
We encourage you to use additional literature in the research on your own ideas. We expect 3-4 additional references, if you want to include more you are welcome to do so. Cite the references in appropriate places, i.e. if you use concepts developed in a different publication or if you want to link to additional information. Your own research idea can be developed from scratch - if you have a great idea, try to sketch it and a possible solution. You can also discuss limitations or ideas that you feel are missing from the paper (You can choose one of the two). A further possibility is to perform research on developments that followed the papers of your topic. If you persue this route, make sure that you do not simply copy parts of other papers. In this case you should base your "own ideas" section on several follow-up publications. A good resource for researching links between articles is Google Scholar. The "cited by" link can sometimes give useful information. Often free links to pdf's can be found as well. Another source for freely downloadable scientific articles is Citeseer. If you can only find a pay link, send us an email - we will try to get you a copy.
Participants are expected to read every paper in preparation for the upcoming presentations. We expect students to actively engage in discussions to further understanding of the presented material. We aim at creating a creative atmosphere - ideas developed during the seminar work might lead to Master thesis projects.
To encourage a more lively atmosphere in the discussion we decided that you can improve your grade by one step, e.g. from 2.0 to 1.7, if you participate actively in the discussions after the talks.
Time: | Fridays 14:00-16:00 |
---|---|
First seminar: | April 15, 2011 |
Room: | Building E1.4 (Max-Planck-Institut), Room 019 |
Registration: | open |
The first presentation by a student is scheduled for May 6 or May 13 2011 (to be discussed), and then every week until the end of the term. The exact schedule will appear here once all topics are selected.
https://lists.mpi-inf.mpg.de/listinfo/cvcg
The details of the seminar and the available subjects will be discussed in the first class on April 15. Attendance is required in all classes. In case you want to participate in the seminar, please register by sending an Email to theobalt@mpii.de or kkim@mpii.de. If possible, also take a moment already to look through the list of topics and provide a ranked list of the three favorite topics you would like to work on: (1) best, (2) second best etc. It will still be possible to provide that list after the first class, but, in any case, please register before the first class via Email. Based on your preference we plan to assign topics by April 19. Please note that the number of available presentation slots is limited. We will try to accommodate all requests for participation but in case we have many more requests than available slots, we will assign slots on a first come first serve basis. Any official announcement will be made through the mailing list. Please subscribe/unsubscribe it.
Christian |
Wedensdyas, 14-15, room 228 |
Kwang In |
Wednesdays, 14-15, room 225 |
Kiran |
Thrusdays, 16-17, room 226 |
Miguel |
Fridays, 16-17, room 221 |
Topic | Authors | Paper Title | Published at | Contact | Presenter | Date | Slides |
---|---|---|---|---|---|---|---|
Motion deblurring |
Anat Levin, Peter Sand, Taeg Sang Cho, Fredo Durand, William T. Freeman Rob Fergus, Barun Singh, Aaron Hertzmann, Sam T. Roweis, William T. Freeman |
SIGGRAPH 2008 SIGGRAPH 2006 |
Kwang In |
Han Du |
May 13 |
|
|
Super-resolution |
Daniel Glasner, Shai Bagon, Michal Irani Michal Irani, Shmuel Peleg |
ICCV 2009 CVGIP 1991 |
Kwang In |
JingYu Yang |
May 20 |
|
|
Space-time super-resolution |
Eli Shechtman, Yaron Caspi, Michal Irani Oded Shahar, Alon Faktor, Michal Irani |
PAMI 2005 CVPR 2011 |
Kwang In |
Madina Boshtayeva |
May 27 |
|
|
Image and video inpainting |
Connelly Barnes, Eli Shechtman, Adam Finkelstein, Dan B. Goldman Yonatan Wexler, Eli Shechtman, Michal Irani |
PatchMatch: A randomized correspondence algorithm for structural image editing |
SIGGRAPH 2009 IEEE Transactions on Pattern Analysis and Machine Intelligence (PAMI), 2007 |
Miguel |
Jing Niu |
Jun 03 |
|
Advance image and video editing |
Connelly Barnes, Dan B. Goldman, Eli Shechtman, Adam Finkelstein Yael Pritch, Eitam Kav-Venaki, Shmuel Peleg |
SIGGRAPH 2010 CVPR 2010 |
Miguel |
Sayyed Auwn Muhammad |
Jun 10 |
|
|
Shape from silhouette |
Wojciech Matusik, Chris Buehler, Ramesh Raskar, Leonard McMillan, Steven Gortler Aldo Laurentini |
The Visual Hull Concept for Silhouette-Based Image Understanding |
SIGGRAPH 2000 IEEE Transactions on Pattern Analysis and Machine Intelligence (PAMI), 1994 |
Kiran |
Rui Xu |
Jun 17 |
|
Reconstruction from community photo collections |
Noah Snavely, Steven M. Seitz, Richard Szeliski Michael Goesele, Noah Snavely, Brian Curless, Steven M. Seitz, Hugues Hoppe |
ACM Transactions on Graphics (Proc. of SIGGRAPH), 2006 Proc. of International Conference on Computer Vision (ICCV), 2009 |
Christian |
Wenkai Dai |
Jul 01 |
|
|
Real-time Motion Capture with Range Sensors |
Jamie Shotton, Andrew Fitzgibbon, Mat Cook, Toby Sharp, Mark Finocchio, Richard Moore, Alex Kipman, and Andrew Blake Robert Y. Wang, Jovan Popović |
Real-Time Human Pose Recognition in Parts from a Single Depth Image |
CVPR 2011 SIGGRAPH 2009 |
Kiran |
Haiyang Xu |
Jul 15 |
|