Generalized Image Acquisition and Analysis

BlurTags: Spatially Varying PSF Estimation with Out-of-Focus Patterns

In this paper, we consider the problem of animation reconstruction, i.e., the reconstruction of shape and motion of a deformable object from dynamic 3D scanner data, without using user provided template models. Unlike pre- vious work that addressed this problem, we do not rely on locally conver- gent optimization but present a system that can handle fast motion, tem- porally disrupted input, and can correctly match objects that disappear for extended time periods in acquisition holes due to occlusion. Our approach is motivated by cartography: We first estimate a few landmark correspon- dences, which are extended to a dense matching and then used to recon- struct geometry and motion. We propose a number of algorithmic building blocks: a scheme for tracking landmarks in temporally coherent and inco- herent data, an algorithm for robust estimation of dense correspondences under topological noise, and the integration of local matching techniques to refine the result. We describe and evaluate the individual components and propose a complete animation reconstruction pipeline based on these ideas. We evaluate our method on a number of standard benchmark data sets and show that we can obtain correct reconstructions in situations where other techniques fail completely or require additional user guidance such as a template model.


Scientific Writing

Block Seminar in winter term 2011/12

Lecturer: Ivo Ihrke

General Information

News: Slides Paper Template

When: 2012, Feb. 27th to Feb. 29th, course starts at 10am (c.t.) every day
Where: E1.7 room 0.01
Registration for mailing list: send email to Ivo Ihrke (


Written communication is the major tool for dissemination of research results and for building a personal standing inside the research community. The ability to communicate clearly is an essential ingredient for successful writing in research. However, there are numerous restrictions due to customs and accepted practices that restrict the ways in which scientists communicate. I will try to give insight into these practices specifically in computer science.

A major goal of the seminar is to get hands-on experience in writing a research report which in many ways resembles a research paper. Therefore, the seminar is project-based, that is, all participants are expected to have a project that they are willing to write about. We assume that the actual research is already done. This implies that all technical details, evaluation experiments, literature search, and comparisons are available.

The course will concentrate on how to organize this material in a best possible manner and how to make a convincing argument for your research. We will cover the major parts of a research paper and how they are used to convey and reinforce your ideas.

The course is practical in nature. Every participant is expected to work on the presentation of their own respective projects. For this it will be necessary to have a material collection upon starting the course. Usually, writing and performing evaluation experiments or generating additional results is a cyclic process where the writing indicates, for example, a lack of convincing evidence for some claims you might want to make. Since the course is short, we can not use this working style. It is thus advisable to start thinking about these issues in advance and to attend the class well prepared.

As a result of the course you will have a clear idea of the structure of your report and set of notes indicating the flow of arguments and a detailed organization of your material. As a final exercise you will have two weeks to complete the report. I will read it and give you feedback. Your grade is determined by active participation in the seminar and by the quality of your final report.


Q: What are the exact dates of the seminar and when is the deadline for the final paper ?
A: The dates are Feb. 27th to 29th. The final write up is due 1st of April.

Q: Will you provide me with a project to write about ?
A: No, you should have a project on your own. This can be a project you are doing in the course of your study or one that you do in your free time.

Q: What are suitable projects for this course ?
A: For examples of projects performed as part of my lecture "Computational Photography" (not necessarily for style of presentation) check "Student project websites" at the bottom of the course webpage.

Q: Is my Open Source project suitable for this course ?
A: Yes, however, you should be prepared to invest some extra thought on how to compare your results/performance/etc. to alternative projects/software. The main difficulty will be the application of scientific standards to your project.

Q: If I do not have a project can I still attend the course ?
A: Yes, however, you will not receive any credits for it. You should also consider that you will benefit most if you are actively working on writing about your own project as compared to only passively absorbing some abstract ideas.

Q: How is the course structured ?
A: The course consists of me lecturing about the rules of scientific writing, style questions, general considerations regarding drafting and revising research papers, etc. in conjunction with you becoming active, working on the presentation of your project. This work will be done individually as well as in groups during the course.

Q: What is the duration of the course ?
A: The course will take three to four days full-time (eight hours per day), plus your work at home when finishing up the writing. The more you can do during the course the less you will have to do at home.

Go back to lecture list