David Lerch is an Artificial Intelligence Researcher with expertise in computer vision, multi-modal representation learning, and semantic situation analysis. At Fraunhofer IOSB, he contributes to the KARLI and SALSA projects, focusing on in-cabin monitoring. He previously worked at KIT and ArtiMinds Robotics on LID comparison, camera pose optimization, and deep learning. David has several years of experience in customer-facing AI projects and his research focuses on multi-modal learning and vision-language models.

At Fraunhofer IOSB 800 researchers work on optics and image recognition. The team perceptual user interfaces works on computer vision for automotive incabin sensing and develops a multi modal foundation model to flexibly answer all possible questions about the cabin interior and passengers. We provide a modular and flexible toolkit to rapidly create proof-of-concept demos and evaluations related to in cabin sensing. We research the latest algorithms for image recognition and evaluate their potential for measuring complex driver states such as cognitive attention, sleep and relaxation and other general situations in the passenger car or truck cabin.