Papers
7
Total Citations
411
H-Index
6
About
Reza Sabzevari’s research lies at the intersection of active perception, 3D reconstruction, and autonomous navigation, with a strong emphasis on enabling robots and vehicles to intelligently interpret their surroundings. His most impactful contributions center on next-best view selection for volumetric 3D object reconstruction, where he pioneered the use of probabilistic information gain metrics to guide a mobile robot’s camera placement. Two foundational papers on this topic—published in 2016 and 2017—have together accumulated over 300 citations, establishing a benchmark for active reconstruction strategies. Sabzevari also made significant advances in multi-body motion estimation, developing methods to simultaneously recover a vehicle’s ego motion and the trajectories of multiple moving objects from a single monocular camera—a critical capability for autonomous driving. Earlier in his career, he explored multisensor data fusion for advanced driver assistance systems and designed intelligent vision systems for robotic manipulation, including a neural-network-based object detection system for a ping-pong playing robot. His work bridges theoretical rigor with practical robotics, offering clear formulations that have been widely adopted by researchers in active perception and autonomous systems.
Research Focus
Key Achievements
Top Papers
- 1
- 2An information gain formulation for active volumetric 3D reconstruction152 citations · 2016
- 3Multi-body Motion Estimation from Monocular Vehicle-Mounted Cameras61 citations · 2016
- 4Multisensor Data Fusion Strategies for Advanced Driver Assistance Systems19 citations · 2009
- 5
- 6Employing ANFIS for Object Detection in Robo-Pong.7 citations · 2008
- 7AN INTELLIGENT VISION SYSTEM ON A MOBILE MANIPULATOR2 citations · 2008