Wieland Morgenstern is a Research Associate at the Computer Vision and Graphics Group in the Vision and Imaging Technologies Department.
News
We won the Research Impact Award for our work on Self-Organizing Gaussians at CVMP 2025!
Our 3DGS.zip compression survey paper has been accepted as a STAR paper (State-of-the-Art Report) and will be presented in a 90-minute session at Eurographics 2025.
Our paper Compact 3D Scene Representation via Self-Organizing Gaussian Grids has been accepted at ECCV 2024 and will be presented in Milano!
We presented our paper Animatable Virtual Humans: Learning pose-dependent human representations in UV space for interactive performance synthesis at IEEE VR 2024!
Our paper "X-maps: Direct Depth Lookup for Event-based Structured Light Systems" has been accepted at the CVPR 2023 Workshop on Event-based Vision. Find more information on the project page https://fraunhoferhhi.github.io/X-maps/
Biography
Wieland Morgenstern is performing research in animation of humans from volumetric video, pairing mesh-based approaches with machine learning, and investigating hybrid model-based Neural Radiance Field (NeRF) approaches. A recent focus is a compact 2D-grid based representation of 3D scenes (including humans) with 3D Gaussian Splatting. Another research area is building Spatial Augmented Reality experiences with event cameras.
Before joining Fraunhofer HHI in 2018, Wieland worked as a Computer Vision Engineer at VideoStitch/Orah from 2014-2018. He received a B.Sc. degree in 2012 and a M.Sc. degree in 2015 from Ilmenau Technical University.
Publications
Student Projects Supervised
Student Projects Supervised
- Alexandra Zimmer: Efficient Sparse Stereo Depth Estimation for Event Cameras, Master Thesis (TU Berlin), 2024.
- Paul Knoll: Posed Neural Radiance Fields for Human Animation, Master Thesis (Johannes Kepler University Linz), 2024.
- Kilian L. Krause: Real-Time Rendering of Dynamic Scenes with Neural Radiance Fields, Bachelor Thesis (TU Berlin), 2024.
- Simon Baumann: 3D Reconstruction with Structured Light using Event-Based Vision and a Video Laser Projector, Master Thesis (TU Berlin), 2022.
Research Projects
LUMINOUS
Vision for Next-Gen XR
Future XR technologies must transcend current limitations to offer deeply realistic and impactful experiences, tailored to understand users’ unique contexts, preferences, and interactions. Such advancements are vital for broadening XR’s application in areas with significant social and economic benefits, like personalized education, daily support, rehabilitation, immersive training, and tele-presence solutions. Achieving this requires XR systems that not only meet but exceed our expectations, presenting responses and interactions that align seamlessly with human perception.
Historically, R&D in XR has focused on spatial user and environment tracking, resulting in static, non-adaptive visual representations. Despite recent strides in AI, including IoT for real-time interaction tracking, machine learning for behavior prediction, and NLP for voice commands, these technologies have been integrated into XR systems as isolated components. The challenge ahead lies in synthesizing these advances into cohesive, intelligent systems that genuinely understand and adapt to individual user needs, setting the stage for a revolution in how we interact with digital environments.
SPIRIT
Scalable Platform for Innovations on Real-time Immersive Telepresence
https://www.spirit-project.eu/
October 2022 - September 2025
VoluProf
Volumetric professor for omnipresent and user-optimized teaching in mixed reality
September 2021 – August 2024
INVICTUS
Innovative Volumetric Capture and Editing Tools for Ubiquitous Storytelling
October 2020 – September 2022
Content4All
Personalised Content Creation for the Deaf Community in a Connected Digital Single Market
September 2017 - November 2020
REPLICATE
cReative-asset harvEsting PipeLine to Inspire Collective-AuThoring and Experimentation
January 2016 – December 2018
Awards
Research Impact Award at CVMP 2025
Our paper Compact 3D Scene Representation via Self-Organising Gaussian Grids received the Research Impact Award at the ACM SIGGRAPH European Conference on Visual Media Production (CVMP) 2025. I led this work as main author, together with Florian Barthel, Anna Hilsmann, and Peter Eisert from Fraunhofer Heinrich-Hertz-Institut HHI and Humboldt-Universität zu Berlin. The award is given to individuals whose key research has been taken up by third parties and used effectively in media production and products.
Innovationspreis für Medientechnologie 2024
The Vision and Imaging Technologies (VIT) department received the Innovation Award for Media Technology in Information Technology from the German Television and Cinema Technology Society (FKTG) at its 30th specialist conference for the development of 3D Human Body Reconstruction (3D HBR) as a core technology for volumetric video capture. The system uses a ring of 16 stereo camera pairs to record real people simultaneously from all directions and reconstruct dynamic 3D models that move naturally and can be viewed from arbitrary virtual viewpoints. I contributed work in temporal mesh registration and mesh-based avatar animation.
CVMP Best Paper Award 2019
The paper "Progressive Non-rigid Registration of Temporal Mesh Sequences" received the Best Paper Award at the 6th ACM SIGGRAPH European Conference on Visual Media Production (CVMP) in London. The prize is sponsored by the ACM Europe Council. Dr.-Ing. Anna Hilsmann, group leader Computer Vision and Graphics at Fraunhofer HHI, and Prof. Dr.-Ing. Peter Eisert, head of the Vision & Imaging Technologies department at Fraunhofer HHI, received the award as co-authors.