BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//Memento EPFL//
BEGIN:VEVENT
SUMMARY:IC Colloquium : Capturing and Editing Models of the Real World in 
 Motion
DTSTART:20171120T161500
DTEND:20171120T173000
DTSTAMP:20260917T092954Z
UID:e9e5618594380c8e7a9eba1cfb9de50187f94a01116b40ad447ce88e
CATEGORIES:Conferences - Seminars
DESCRIPTION: By : Christian Theobalt - Max Planck Institute for Informati
 cs\n\nAbstract :\nNew methods for capturing highly detailed models of movi
 ng real world scenes with cameras\, i.e.\, models of detailed deforming ge
 ometry\, appearance or even material properties\, become more and more imp
 ortant in many application areas. They are needed in visual content creati
 on\, for instance in visual effects\, where they are needed to build highl
 y realistic models of virtual human actors. Further on\, efficient\, relia
 ble and highly accurate dynamic scene reconstruction is nowadays an import
 ant prerequisite for many other application domains\, such as:\nhuman-comp
 uter and human-robot interaction\, autonomous robotics and autonomous driv
 ing\, virtual and augmented reality\, 3D and free-viewpoint TV\, immersive
  telepresence\, and even video editing. The development of dynamic scene r
 econstruction methods has been a long standing challenge in computer graph
 ics and computer vision. Recently\, the field has seen important progress.
  New methods were developed that capture - without markers or scene instru
 mentation - rather detailed models of individual moving humans or general 
 deforming surfaces from video recordings\, and capture even simple models 
 of appearance and lighting. However\, despite this recent progress\, the f
 ield is still at an early stage\, and current technology is still starkly 
 constrained in many ways. Many of today's state-of-the-art methods are sti
 ll niche solutions that are designed to work under very constrained condit
 ions\, for instance: only in controlled studios\, with many cameras\, for 
 very specific object types\, for very simple types of motion and deformati
 on\, or at processing speeds far from real-time. In this talk\, I will pre
 sent some of our recent works on detailed marker-less dynamic scene recons
 truction and performance capture in which we advanced the capabilities of 
 existing approaches in several ways. I will show new methods for marker-le
 ss skeletal motion and performance capture in less constrained environment
 s\, even in outdoor scenes\, and with a low number of cameras. I will also
  discuss new state-of-the-art methods for marker-less motion capture of ha
 nds\, also when in interaction with objects and background clutter\, from 
 a single RGB-D camera. Finally\, I will show new approaches for high-quali
 ty capture of deforming surfaces and the human face from monocular color a
 nd depth cameras\, even in real-time. If time allows\, I will also briefly
  show how dynamic scene reconstruction methods may be used for advanced vi
 deo editing effects and how joining forces of machine learning and model-b
 ased methods may open up new possibilities.\n\nBio :\nChristian Theobalt i
 s a Professor of Computer Science and the head of the research group "Grap
 hics\, Vision\, and Video" at the Max-Planck-Institute (MPI) for Informati
 cs\, Saarbruecken\, Germany. Most of his research deals with algorithmic p
 roblems at the intersection of Computer Graphics\, Computer Vision\, HCI a
 nd Machine Learning\, such as:\nstatic and dynamic 3D scene reconstruction
 \, marker-less motion and performance capture\, computer animation\, virtu
 al and augmented reality\, appearance and reflectance modeling\, machine l
 earning for graphics and vision\, new sensors for 3D acquisition\, advance
 d video processing\, as well as image- and physically-based rendering. For
  his work\, he received several awards\, including the Otto Hahn Medal of 
 the Max-Planck Society in 2007\, the EUROGRAPHICS Young Researcher Award i
 n 2009\, and the German Pattern Recognition Award 2012. In 2015 he was ele
 cted one of the top 40 innovation leaders under the age of 40 in Germany b
 y the magazine Capital. Further\, in 2013 he was awarded an ERC Starting G
 rant by the European Union\, the most prestigious and most competitive gra
 nt for individual researchers. He is a co-founder of a spin-off company fr
 om his group - www.thecaptury.com - that sells the most advanced marker-le
 ss motion and performance capture solution commercially available today.\n
 \nMore information\n 
LOCATION:BC 420 https://plan.epfl.ch/?room==BC%20420
STATUS:CANCELLED
END:VEVENT
END:VCALENDAR