PamiTC Job Board - Posting Details

Title: PhD Position: Deep-Learning-based Dynamic Scene Reconstruction in General Environments Posted: June 12, 2017
Company/Institution: Max-Planck-Institute (MPI) for Informatics
Location: Saarbruecken, Germany

Description:

I am looking for a PhD student to join my research group Graphics, Vision and Video at the Max-Planck-Institute (MPI) for Informatics in Saarbruecken, Germany. The position can be filled immediately, but starting date is felxible.

The student will work with me on a topic in the general area of real-time dynamic scene reconstruction with only one or two cameras. Dynamic scene reconstruction is a profoundly hard problem at the intersection of vision and graphics which is of fundamental relevance to intelligent, interactive and autonomous computing systems of the future. Future autonomous robots and vehicles, future virtual and augmented reality systems, and future telepresence systems will fundamentally depend on a new generation of methods that capture high quality models of the real world in motion in a scalable, robust and fast way in general uncontrolled scenes. The Graphics, Vision and Video Group has pioneered methods in off-line and real-time marker-less motion and performance capture of humans, faces, hands and deformable objects, and has pioneered ways of efficient and high-quality inverse rendering. Very recent examples are the Face2Face face reenactment approach from CVPR 2016 that also won the 'Best of Show Award' at Emerging Technologies of SIGGRAPH 2016, the VNect real-time monocular full-body motion capture approach (SIGGRAPH 2017), and the first approach for real-time intrinsic video decomposition (SIGGRAPH 2016).

Take a look at our list of projects and publications:
http://gvv.mpi-inf.mpg.de/GVV_Projects.html
http://gvv.mpi-inf.mpg.de/GVV_Publications.html

Some of our research results also form the basis for our award-winning startup company the Captury that commercializes the most advanced real-time marker-less motion capture technology that is available on the market.

As a PhD student, you will join a vibrant and dynamic research environment in which we constantly push the boundary of what is possible today in visual computing. The MPI for Informatics has a wide range of state-of-the-art computing and research equipment, including several large scale 3D / 4D scanners and a full multi-view motion and performance capture studio featuring several multi-camera and multi-sensor systems. You will rethink the basic algorithmic concepts of marker-less motion capture and dynamic(4D) scene reconstruction to make them scalable to real-world complexity and diversity, to real-time performance, and to make them work with simple lightweight sensors, e.g., just a single RGB camera. In this context, we will also investigate profound methodical integration of deep learning-based and conceptually advanced model-based real world reconstruction techniques.


Application Instructions:

Applicants should have a Bachelor, preferably Master degree, in computer science or a closely related field. Course work and research experience in one or more of the following areas is desirable: multi-view / 3D scene reconstruction methods, computer graphics, 4D reconstruction, machine learning for visual computing, computer animation, marker-less performance capture, marker-less motion capture, tracking algorithms, 3D Video. Applicants should have an excellent academic track record; publications in one of the above areas are a plus. A candidate should be fluent in written and spoken English and be willing to travel. Full funding and benefits are provided.

If you are interested in this PhD position, please send a complete application package, including a CV, a research statement, transcripts and certificates, and the contacts of two references by

>>>>>>> June 30 2017 <<<<<<<<<<

to
Prof. Dr. Christian Theobalt
MPI for Informatics
theobalt@mpi-inf.mpg.de

please add the tag [PhD Application] in the subject of your Email.

About Christian Theobalt and the GVV group

Christian Theobalt is a Professor of Computer Science and the head of the research group "Graphics, Vision and Video" at the Max-Planck-Institute (MPI) for Informatics, Saarbruecken, Germany. Most of his research deals with algorithmic problems that lie on the boundary between the fields of Computer Vision and Computer Graphics, such as static and dynamic 3D scene reconstruction, marker-less motion capture, virtual and augmented reality, computer animation, appearance and reflectance modeling, machine learning for graphics and vision, new sensors for 3D acquisition, advanced video processing, as well as image- and physically-based rendering. He has published more than 60 papers in the top vision and graphics conferences. For his work, he received several awards, including the Otto Hahn Medal of the Max-Planck Society in 2007, the EUROGRAPHICS Young Researcher Award in 2009, and the German Pattern Recognition Award 2012. In 2015 he was elected one of the top 40 innovation leaders under the age of 40 in Germany by the magazine Capital. Further, in 2013 he was awarded an ERC Starting Grant by the European Union, the most prestigious and most competitive grant for individual researchers. He is a Principal Investigator and a member of the Steering Committee of the Intel Visual Computing Institute in Saarbruecken. He is also a co-founder of a spin-off company from his group - www.thecaptury.com - that is commercializing the most advanced marker-less motion and performance capture solution commercially available today.

Several of our recent projects received attention in popular press, e.g.
http://www.theverge.com/2015/10/15/9541985/Real-time-expression-transfer-facial-reenactment

The group Graphics, Vision & Video has close collaborations with international academic and industry partners and lives a very collaborative team-oriented working style within the group itself. Check out our team (link)

About the environment

The Max-Planck Institute for Informatics (MPI-INF) (www.mpi-inf.mpg.de) is one of the world's leading research institutes in Computer Science in general, and Visual Computing in particular. It is located on the campus of Saarland University in Saarbruecken, Germany. MPI-INF is embedded in a unique cluster of computer science research. Around 400 PhD students in CS do research in the different CS institutes on campus under the roof of a joint CS graduate school. In immediate neighborhood on campus, there are several other computer science research institutes of world renown with which close collaborations exist: the German Research Center for Artificial Intelligence (DFKI), the Max-Planck-Institute for Software Systems, the Institute for Bioinformatics, the Excellence Cluster Multimodal Computing and Interaction, the new federal research center on IT Securiy, Privacy and Accountability (CISPA), and the Computer Science Department of Saarland University. The Leibniz Center for Informatics in Schloss Dagstuhl is also located nearby. The Intel Visual Computing Institute (IVCI) on campus further strenghtens the visual computing research focus in Saarbruecken. (www.informatik-saarland.de)