Omid Hosseini Jafari

RWTH Aachen University, TU Dresden, Torc Robotics (United States)

Papers

Total Citations

200

H-Index

About

Omid Hosseini Jafari is a computer vision and robotics researcher whose work focuses on real-time perception systems for autonomous platforms. His primary research areas include RGB-D based people detection and tracking, pedestrian detection, and stereo vision for mobile robots and wearable cameras. Jafari's most influential contribution is his 2014 paper on real-time RGB-D based multi-person detection and tracking, which has garnered 181 citations. This work integrates visual odometry, ground plane estimation, and multi-hypothesis tracking to enable robust people detection for mobile robots and head-worn cameras—a critical capability for human-robot interaction and assistive technologies. He further advanced the field with a depth-based template matching approach for pedestrian detection (2016, 16 citations), addressing challenges in multi-person scenarios. Most recently, his 2024 work on flow-guided online stereo rectification tackles the practical problem of maintaining stereo calibration in autonomous vehicles and robots operating in dynamic environments, where vibration and structural stress cause constant recalibration needs. Jafari's research bridges the gap between laboratory algorithms and real-world deployment, making him a notable contributor to the development of reliable perception systems for autonomous navigation and human-aware robotics.