Hello and welcome to my homepage. My name is Holger Caesar and I am currently a Senior Research Scientist at Motional (formerly nuTonomy) in Singapore. Working on the Machine Learning Team, my job is to make autonomous vehicles perceive and understand their environment.

I previously did my PhD in Computer Vision under the supervision of Prof. Vittorio Ferrari at the University of Edinburgh and ETH Zurich. Before that I studied at KIT and EPFL Lausanne.
During the course of my PhD I have proposed two state-of-the-art semantic segmentation methods (joint calibration, end-to-end), analyzed thing and stuff classes in context and created the popular COCO-Stuff dataset. Furthermore I was a reviewer for various journals and conferences (incl. ECCV, BMVC, CHI, VCIP, MULT-D) and created the largest existing lists of Semantic Segmentation and Generative Adversarial Network (GAN) papers.
I have worked on many different projects, including the analysis of aviation accidents, online marketing, middleware messaging in banks and communication in the energy sector. My academic journey has taken me to software design, virtual reality, speech recognition, computer graphics and simulation, hand-gesture recognition, activity recognition and semantic segmentation.
I grew up in the Southwest of Germany along the beautiful Hessian Mountain Road. As such, when I am not working or singing, I am probably hiking or biking in the local mountains.