The Computer Vision and Graphics (CVG) group is an active research group carrying out innovative research in the field of Visual Computing. We combine extensive expertise in Computer Vision on the analysis side, i.e. capturing, sensing and understanding the visual real world and Computer Graphics on the synthesis side, i.e. modelling into digital representations and animation of these models as well as rendering and visualization, often in an analysis-by-synthesis cycle. In this context, we combine CV & CG models with deep neural networks and especially develop model-based deep learning methods, that integrate a-priori and model knowlegde into neural architectures. We develop sophisticated solutions for virtual, augmented and extended reality for a broad range of applications for industry, multimedia, medicine and security. Current topics include human shape/motion analysis and synthesis, computational video, image and video understanding, biomedical image analysis, generative models as well as augmented reality.
Latest Publications
- September 2023: Wolfgang Paier presented our paper "Video-Driven Animation of Neural Head Avatars" at VMV 2023
- September 2023: We presented 5 papers at BMT 2023
- From Multispectral-Stereo to Intraoperative Hyperspectral Imaging: a Feasibility Study
- 3D Hyperspectral Light-Field Imaging: a first Intraoperative Implementation
- Automatic Registration of Anatomical Structures of Stereo-Endoscopic Point Clouds
- Surgical Phase Recognition for Different Hospitals
- Modified Allen Test Assessment via Imaging Photoplethysmography
- July 2023: Aleixo Cambeiro Barreiro presented our paper "Automatic Reconstruction of Semantic 3D Models from 2D Floor Plans" at the Int. Conf. on Machine Vision and Applications 2023.
- June 2023: Arian Beckmann presented our paper "Fooling State-of-the-Art Deepfake Detection with High-Quality Deepfakes"at the ACM Workshop on Information Hiding and Multimedia Security.
- June 2023: Wieland Morgenstern presented our paper "X-maps: Direct Depth Lookup for Event-based Structured Light Systems" at the CVPR 2023 Workshop on Event-based Vision. Check out the project page as well as the code on GitHub
- May 2023: Jannes Magnusson presendet our paper "Towards L-System Captioning for Tree Reconstruction" at Eurographics 2023 in Saarbrücken, Germany.
- Januar 2023: The source code to our paper "CASAPose: Class-Adaptive and Semantic-Aware Multi-Object Pose Estimation" is available on GitHub.
Research Topics
The research addresses different fields in the area of Computer Vision, Computer Graphics and Visual Computing.
Projects
Current research at the CVG group is based on numerous projects from industry and public funding bodies on European and national level.
Publications
Journal papers, conference proceedings, talks and tutorials, standardization contributions and books. Find out about the publications of our group.
Team
Your partner for research and product development: Get in contact with our scientists.

Student Opportunities
We offer exciting topics for Bachelor- and Masterthesis or opportunities to work with us as a student research assistant.