Research Highlights: Prof. Niessner, Dr. Thies, and M.Sc. Rössler presenting to the German Federal Government at #Digitalgipfel about #AI, #SyntheticMedia, and #MediaForensics. A specific focus was put on the FaceForensics works, as well as a general discussion about the general AI landscape in Germany. So many opportunities in how to move the country forward!

Research Highlights: ScanNet is an RGB-D video dataset containing 2.5 million views in more than 1500 scans, annotated with 3D camera poses, surface reconstructions, and instance-level semantic segmentations. This data is specifically designed to support deep learning tasks for scene understanding in 3D - it is one of the largest annotated RGB-D dataset that is publically available.

Research Highlights: Group Retreat of our TUM Visual Computing & AI Group! It was a fantastic venue at Beilngries next to the idyllic Altmuehl river. We had a fantastic time going through paper drafts, presentations of our ongoing projects -- made a lot of progress -- amazing group!

Research Highlights: Face2Face is a real-time face tracker whose analysis-by-synthesis approach preceisely fits a 3D face model to a captured RGB video. This produces high accuracy tracking, allowing for photo-realistic re-rendering and modifications of a target video: in a nutshell, one can change the expressions of a target video in real time. This project has received incredible attention with several million youtube views and wide range of media coverage. We even gave live demos on several occasions on public television!

The Visual Computing & Artificial Intelligence Lab at TUM, is a group of research enthusiasts advancing cutting-edge research at the intersection of computer vision, computer graphics, and artificial intelligence. Our research mission is to obtain highly-realistic digital replica of the real world, which include representations of detailed 3D geometries, surface textures, and material definitions of both static and dynamic scene environments. In our research, we heavily build on advances in modern machine learning, and develop novel methods that enable us to learn strong priors to fuel 3D reconstruction techniques. Ultimately, we aim to obtain holographic representations that are visually indistinguishable from the real world, ideally captured from a simple webcam or mobile phone. We believe this is a critical component in facilitating immersive augmented and virtual reality applications, and will have a substantial positive impact in modern digital societies.

Keywords: Visual Computing, 3D Vision, Machine Learning, Artificial Intelligence, 3D Reconstruction, Scene Understanding, Neural Rendering, Neural Scene Representations

Our lab is located at the research campus in Garching. Inside the Faculty of Computer Science, the slide enables both students and visitors to make the journey down from the fourth floor a little quicker than might be expected.