Computer Graphics
TU Braunschweig

Events


Talk MA Talk: Facial Texture Generation from Uncontrolled Monocular Video

01.09.2017 13:00
Informatikzentrum, Seminarraum G30

Speaker(s): Rudolf Martin

Talk Real Virtual Humans

10.07.2017 13:30
IZ G30

Speaker(s): Gerard Pons-Moll

For man-machine interaction it is crucial to develop models of humans that look and move indistinguishably from real humans. Such virtual humans will be key for application areas such as computer vision, medicine and psychology, virtual and augmented reality and special effects in movies. 

Currently, digital models typically lack realistic soft tissue and clothing dynamics or require time-consuming manual editing of physical simulation parameters. Our hypothesis is that better and more realistic models of humans and clothing can be learned directly from real measurements coming from 4D scans, images and depth and inertial sensors. We combine statistical machine learning techniques and physics based simulation to create realistic models from data.

I will give an overview of several of our projects in which we build realistic models of human pose and shape, soft-tissue dynamics and clothing. I will also present a recent technique we have developed to capture human movement from only 6 inertial sensors attached at the body limbs. This will enable capturing human motion of every day activities, for example while we are interacting with other people, while we are riding a bike or driving a car. Such recorded motions will be key to learn models that replicate human behaviour. I will conclude the talk outlining the next challenges to build virtual humans that are indistinguishable from real people.

Bio: 

Gerard Pons-Moll obtained his degree in superior Telecommunications Engineering from the Technical University of Catalonia (UPC) in 2008. From 2007 to 2008 he was at Northeastern University in Boston USA with a fellowship from the Vodafone foundation conducting research on medical image analysis. He received his Ph.D. degree (with distinction) from the Leibniz University of Hannover in 2014. In 2012 he was a visiting researcher at the vision group at the University of Toronto. In 2012 he also worked as intern at the computer vision group at Microsoft Research Cambridge. From 11/2013 until 11/2015 he was a postdoc at the Max Planck Institute (MPI) for Intelligent Systems in Tuebingen, Germany. Since 11/2015 he is a research scientist at the MPI.

His work has been published in the major computer vision and computer graphics conferences and journals including Siggraph, Siggraph Asia, CVPR, ICCV, BMVC(Best Paper), Eurographics(Best Paper), IJCV and TPAMI. He serves regularly as a reviewer for TPAMI, IJCV, Siggraph, Siggraph Asia, CVPR, ICCV, ECCV, ACCV and others. He co-organized 3 tutorials at major conferences: 1 tutorial at ICCV 2011 on Looking at People: Model Based Pose Estimation, and 2 tutorials at ICCV 2015 and Siggraph 2016 on Modeling Human Bodies in Motion.

His research interests are 3D modeling of humans and clothing in motion and using machine learning and graphics models to solve vision problems.

Talk Bildbasiertes Messen und Modellieren der realen Welt

20.06.2017 14:00
PTB-Braunschweig, Seminarzentrum A, Kohlrausch-Bau

Speaker(s): Marcus Magnor

Bilder sind Projektionen der physikalischen Realität: sämtliche leuchtenden und beleuchteten Dinge senden kontinuierlich Bilder von sich aus, in alle Richtungen und über weite Distanzen. Mit Lichtgeschwindigkeit übertragen Bilder reichhaltige Information über ihren Entstehungsort und ihre Entstehungsweise. Damit stellt jedes digitale Bild eine Messung dar, jede Digitalkamera ist ein Messinstrument, das simultan Millionen von Messwerten unserer natürlichen Umgebung erfasst, aus der Distanz und ohne das vermessene System zu beeinflussen.

In meinem Vortrag möchte ich anhand von Beispielen aus der Radioastronomie, Strömungsmechanik, Atmosphärenoptik und Wahrnehmungspsychologie zeigen, wie moderne Bildverarbeitung und Rechnersimulation in der Lage ist, aus Bildern komplexer natürlicher Vorgänge quantitative Informationen zu ermitteln.

[Weitere Informationen]

Symposium SVCP17 - Symposium on Visual Computing and Perception

07.06.2017 13:00 - 08.06.2017 16:00
Campus Nord, TU Braunschweig

Speaker(s): Micheal Bach, Heinrich Bülthoff, Jan Koenderink, Holger Theisel, Bernt Schiele, Erik Reinhard

Computer graphics, vision, and psychophysics form a scientific triad whose interdisciplinary research is about to fundamentally change the way we drive our cars, watch movies, play games, communicate with one another and computers, socially interact, learn, live. In six distinguished lectures, some of the most renowned scientists worldwide in these fields will share their view on what has already been accomplished, which challenges still lie ahead, and what we may expect in the future. The symposium is intended to bring together all scientific communities interested in visual computing and perception, to stir everyone's imagination, and to foster exciting new research between graphics, vision, and perception.

SVCP17 - Symposium on Visual Computing and Perception

Talk Einsatz der Unity-Engine in VR-Umgebungen

12.05.2017 13:00
ICG Lab Campus Nord

Speaker(s): Marcus Riemer

Die Unity-Engine ist eine populäre Entwicklungsumgebung für Spiele und interaktive Anwendungen. Neben klassischen Zielplattformen wie PC oder Smartphone bietet Unity auch Entwicklungsschnittstellen für verschiedenste VR Systeme. Marcus Riemer berichtet über den Einsatz von Unity an der FH Wedel sowie die Herausforderungen und Möglichkeiten der Nutzung für individuelle VR Systeme am Beispiel eines CAVE Virtual Environment.

Weitere Vortragende: Steffen Kurt, Florian Habib

Talk Physik-basierte photorealistische Visualisierung astronomischer Nebel

09.12.2016 13:00
IZ G30

Speaker(s): Wolfgang Steffen

Talk Visual Computing - Bridging Real and Digital Domain

06.12.2016 13:00
IZ G30

Speaker(s): Marcus Magnor

Any luminous or illuminated object continuously emits images of itself, in all directions, and over huge distances. At the speed of light images convey a wealth of information about their origins which our visual system can decipher almost without effort, in real time, extremely efficient, and enormously robust. Consequently, the visual sense has evolved to become our prime modality for gathering information of our environs. In my talk I will outline how we have only just begun to exploit visual information as the ideal interface between the real and digital world, as well as between the digital domain and human brains. Visual Computing constitutes a fundamental Enabling Technology of the 21. century that paves the way for various technological advances of substantial economic as well as social impact.

Talk Disputation

14.11.2016 13:00
Informatikzentrum, IZ 161

Speaker(s): Michael Stengel

Gaze-contingent Computer Graphics

Talk Disputation

07.10.2016 13:00
Informatikzentrum, Seminarraum G04

Speaker(s): Thomas Neumann

Reconstruction, Analysis, and Editing of dynamically deforming 3D-Surfaces

Talk Fast image reconstruction for Magnetic-Particle-Imaging by Chebyshev Transformations

08.08.2016 13:00
Informatikzentrum, Seminarraum G30

Speaker(s): Leonard Schmiester

Image reconstruction in magnetic-particle-imaging (MPI) is done using an algebraic approach for Lissajous type measurement sequences. By solving a large linear system of equations the spatial distribution of magnetic nanoparticles can be determined. Despite the use of iterative solvers that converge rapidly, the size of the MPI system matrix leads to reconstruction times that are typically much longer than the actual data acquisition time. For this reason, matrix compression techniques have been introduced that transform the MPI system matrix into a sparse domain and then utilize this sparsity for accelerated reconstruction. Within this scope we investigate the Chebyshev transformation for matrix compression. By reducing the number of coefficients per matrix row to one, it is even possible to derive a direct reconstruction method that circumvents the usage of iterative solvers.

Talk MA-Talk: Guided Camera Placement for Image-Based Rendering

11.07.2016 13:00
Informatikzentrum, Seminarraum G30

Speaker(s): Leslie Wöhler

Talk Promotions-V-Vg: Gaze-Contingent Perceptual Rendering in Computer Graphics

01.07.2016 13:00
Informatikzentrum, Seminarraum G30

Speaker(s): Michael Stengel

Contemporary digital displays feature multi-million pixels at ever-increasing refresh rates. Reality, on the other hand, provides us with a view of the world that is continuous in space and in time. The discrepancy between viewing the physical world and its sampled depiction on digital displays gives rise to perceptual quality degradation. By measuring or estimating where we look, gaze-contingent algorithms aim at exploiting the way we visually perceive to remedy visible artifacts. In his dissertation pre-talk Michael Stengel will present recent results from projects in the field of gaze-contingent and perceptual algorithms. Two projects aim at boosting the perceived visual quality of conventional video footage when viewed on commodity monitors or projection. In addition he will describe a novel head-mounted display with real-time gaze tracking resulting in many novel applications in the context of Virtual Reality and Augmented Reality. In a follow-up project Michael and colleagues derived a novel gaze-contingent render method using active gaze tracking to reduce computational efforts when shading virtual worlds.

Talk Promotions-V-Vg: Reconstruction, Analysis and Editing of Dynamically Deforming 3D Surfaces

10.06.2016 13:00
Informatikzentrum, Seminarraum G30

Speaker(s): Thomas Neumann

Dynamically deforming 3D surfaces play a major role in computer graphics. However, producing time-varying dynamic geometry at ever increasing detail is a labor-intensive process, and so a recent trend is to capture geometry data directly from the real world. The first part of this talk presents novel approaches in this direction, approaches that capture dense dynamic 3D surfaces from multi-camera systems in a particularly robust and accurate way. This provides highly realistic dynamic surface models for phenomena like moving garments and bulging muscles.

However, conveniently re-using, editing or otherwise analyzing dynamic 3D surface data is not yet possible. The second part of the talk thus deals with novel data-driven modeling and animation approaches. I first show a supervised data-driven approach for modeling human muscle deformations, an approach that scales to huge datasets and provides fine-scale, anatomically realistic deformations at a high quality not shown by previous data-driven methods. I then extend data-driven modeling to the unsupervised setting, thus providing editing tools for a wider set of input data ranging from facial performance captures and full body motion to muscle and cloth deformations. To this end, I introduce the concepts of sparsity and locality within a mathematical optimization framework. I also explore these concepts for constructing shape-aware functions that are useful for static geometry processing, registration and localized editing.

Talk MA-Talk: Interactive Realtime Image Segmentation

29.04.2016 11:00
Informatikzentrum, Seminarraum G30

Speaker(s): Moritz Mühlhausen

Talk MA-Talk: Globally Weighted 3D Reconstruction with Accurate Visibility Computation

25.04.2016 13:00
Informatikzentrum, Seminarraum G30

Speaker(s): Marc Kassubeck

Talk Context-aware dynamic Sensor Fusion

28.09.2015 13:00
Informatikzentrum, Seminarraum G30

Speaker(s): Thiemo Alldieck

Sensor fusion aims to compensate for individual strengths and weaknesses of different sensors. When recording a complex scene over a longer period, these characteristics might change, resulting in possible system instabilities. The presented work introduces a new dynamic method for data fusion of different imaging devices, namely RGB and thermal camera. Therefor background conformity values of two image sources are fused in order to enable stable background subtraction for persistent surveillance. Image quality heuristics based on image characteristics and contextual information are specified to evaluate the usefulness of the modalities and perform the fusion context aware.

Talk Three Geometric Structures and their Applications

18.09.2015 13:00
Informatikzentrum, Seminarraum G30

Speaker(s): Nabil Mustafa

Since the beginning of systematic research on geometric computing almost forty years ago, there has been a very fruitful interplay between the mathematical study of geometric structures and the search for efficient and practical solutions for a variety of problems involving geometric data. In this talk I will illustrate this with applications to three different areas: computer graphics, algorithms and combinatorics.

Talk Disputation

07.08.2015 10:00
Informatikzentrum, IZ 161

Speaker(s): Pablo Bauszat

Advanced Denoising and Memoryless Acceleration for Realistic Image Synthesis

Talk Disputation

27.07.2015 14:00
Informatikzentrum, Seminarraum G04

Speaker(s): Benjamin Meyer

Measuring, modeling and simulating the re-adaptation process of the Human Visual System after short-time glares in traffic scenarios

Talk Disputation

17.07.2015 15:00
Informatikzentrum, IZ 161

Speaker(s): Kai Ruhl

Interactive Spacetime Reconstruction in Computer Graphics

Talk Disputation

13.07.2015 13:15
Informatikzentrum, Seminarraum G04

Speaker(s): Maryam Mustafa

ElectroEncephaloGraphics - a Novel Modality for Graphics Research

Talk Disputation

03.07.2015 10:00
Informatikzentrum, Seminarraum G04

Speaker(s): Lorenz Rogge

Augmenting People in Monocular Video Data

Talk Methods for Analyzing the Influence of Molecular Dynamics on Neuronal Activity

26.06.2015 13:00
Informatikzentrum, Seminarraum G30

Speaker(s): Stefan Sokoll

Investigating the functioning of neurons at the molecular level is an important foundation to understand how higher brain functions like perception, behavior, or learning and memory are accomplished. Since molecular processes occur in the nanometer range and have to be studied in living samples, recently developed optical super-resolution techniques have boosted their characterization. However, super-resolution techniques require complex instrumentation, are hardly applicable to organotypic samples and still suffer from relatively low temporal resolution. This talk presents new analysis tools that aim to overcome these limitations and allow to study how the dynamics and the interplay of molecules modulate synaptic transmission efficiency. At first, an approach for the detection of individual presynaptic activity will be briefly introduced, but the major part focuses on an algorithm that facilitates fast 3D molecular dynamic analyses within brain slices. It adjusts astigmatism-based 3D single-particle tracking (SPT) techniques to depth-dependent optical aberrations induced by the refractive index mismatch so that they are applicable to complex samples. In contrast to existing techniques, the presented online calibration method determines the aberration directly from the acquired 2D image stream by exploiting the inherent particle movement and the redundancy introduced by the astigmatism. The method improves the positioning by reducing the systematic errors introduced by the aberrations and allows to correctly derive the cellular morphology and molecular diffusion parameters in 3D independently of the imaging depth.

Talk Promotions-V-Vg: Real-World Video Processing Using Unstructured Scene Representations

26.06.2015 10:00
Informatikzentrum, Seminarraum G30

Speaker(s): Felix Klose

When processing single or multi-view video data recorded in uncontrolled environments using scene reconstruction algorithms a multitude of factors can negatively influence the result quality. These factors include camera, lens or color miscalibrations, errors in temporal or spatial camera alignment, unsynchronized and rolling shutters on the camera side, as well as specular, untextured, repetitive objects or objects with visually complex appearances inside the scene. These circumstances make working with computer vision algorithms on real-world data a very challenging task and errors in measurements in real-world recoding setups can not be avoided and have to be accounted for.

In this talk I will give an overview of my work in single and multi-view video processing of real world data using unstructured scene representations. I show how dense 2D correspondence based stereoscopic free-viewpoint video can be created, using tools for user guided error correction. How the complexity of real-world multi-view data can be handled by tracking small surface patches and using a strict motion model to resolve ambiguities and create quasi dense scene

representations. And finally how to create high quality video effects that can handle extreme amounts of noise in estimated depth maps by leveraging the redundancy inherent in video data.

Talk MA-Talk: Compressed Sensing-based Progressive Reconstruction for Image Synthesis

11.05.2015 13:00
Informatikzentrum, Seminarraum G30

Speaker(s): Cong Wang

Compressed Sensing (CS) is a new mathematical framework for reconstruction of signals with missing information. Recently, its application to sparse image reconstruction, reconstruction of an image from a small set of known pixels, has shown promising results. The key idea is derived from the fact that most natural images are highly compressible because they are sparse in a transform domain. This leads to the obvious questions: Why waste resources on evaluating information (here, individual pixels) that are discarded later on or have only small impact on the overall visual impression? So far the measurements (evaluated pixels of the image) are chosen in random fashion (usually based on a Blue Noise distribution) to uniformly cover the image domain. Theoretically, if salient features of the image are known in advance, fewer measurements would be needed for high-quality reconstruction. For real-world images taken by a photo or video camera it is very hard to evaluate important features of the image without actually capturing them. However, during image synthesis more knowledge about the scene, camera and lighting situation is present. If carefully observed, the rendering process can potentially provide useful cues which are more efficient to evaluate than the actual measurements, can guide the image sampling process, and thus accelerate convergence.