Benjamin Ranft

Karlsruhe Institute of Technology, Center for Information Technology

Papers

Total Citations

H-Index

About

Benjamin Ranft is a researcher whose work sits at the intersection of autonomous perception and real-time embedded computing. His primary contributions lie in dynamic scene understanding for mobile robotics, particularly through visual-based detection and tracking of moving objects in complex urban environments. His most influential work, "Detection and tracking of independently moving objects in urban environments" (2010), with 21 citations, established a stereo-vision-only approach that enables autonomous vehicles to perceive dynamic surroundings without relying on expensive LIDAR—a foundational capability for self-driving cars operating indoors or outdoors. Ranft also made notable strides in adaptive computing for robotics, as demonstrated in his 2012 paper on run-time adaptation to heterogeneous processing units for real-time stereo vision. This work addressed the critical challenge of efficiently distributing computational loads across CPUs and GPUs in modern systems, from smartphones to workstations, ensuring that vision algorithms meet strict real-time deadlines. By bridging perception algorithms with hardware-aware optimization, Ranft’s research has helped pave the way for practical, cost-effective autonomous systems that can operate reliably in dynamic, real-world settings.