Computer Vision
About
At the Computer Vision Team of the CAMP Chair, our research focuses on advancing 3D computer vision, with an emphasis on scene understanding, 3D object recognition, and 3D reconstruction. We aim to build intelligent systems that perceive and understand the world from visual and spatial data, combining deep learning, geometric reasoning, and multimodal modeling.
Scene Understanding
We develop methods that enable machines to understand complex environments at a semantic level. Our work includes:
- Vision-language models for grounding and semantic understanding
- Scene graphs and relational reasoning
- Depth and geometry estimation
- Holistic 3D scene interpretation from images, videos, and point clouds
3D Object Recognition
We study algorithms for recognizing and localizing objects in 3D space, focusing on:
- 3D object detection
- 6D pose estimation
- Learning-based approaches for understanding shape, category, and function
3D Reconstruction
We explore both classical and neural methods to reconstruct high-quality 3D representations from visual input. Topics include:
- Neural Radiance Fields (NeRF)
- Gaussian Splatting and point-based rendering
- Signed Distance Functions (SDFs) and volumetric models
- Real-time and robust SLAM systems
Our research is applied in domains such as robotic perception, augmented reality, autonomous driving, and healthcare, where reliable 3D understanding is essential. We actively collaborate with international research institutions and industry partners to develop technology that bridges fundamental research and real-world applications.
Contact Persons
Group Members
| PD Dr. Federico Tombari | Nikolas Brasch | Dr. Stefano Gasperini |
| Mahdi Saleh | Shun-Cheng Wu | Dr. Benjamin Busam |
| Artem Savkin | Nils Morbitzer | Sen Wang |
| Guangyao Zhai | Markus Herb | Klara Reichard |
| Yan Di | Pengyuan Wang | Hyun Jun Jung |
| Alexander Lehner | Changxuan Li | Christian Kapeller |
| Evin Pınar Örnek | Lennart Bastian | Felix Tristram |
| Kunyi Li |