R. Nevatia

University of Southern California, Texas Instruments (United States)

Papers

Total Citations

H-Index

About

R. Nevatia is a pioneering figure in computer vision, with foundational contributions to scene understanding, motion analysis, and object recognition for autonomous navigation. His research spans dynamic-scene interpretation, generic object localization, and segmentation from imperfect data—areas critical to enabling machines to perceive and navigate real-world environments. In his highly cited 2002 work, Nevatia introduced an efficient method for recognizing and localizing generic objects like desks and doors as landmarks for indoor robot navigation, demonstrating robust performance on real scenes. This work, with 33 citations, exemplifies his ability to bridge theoretical advances with practical robotics applications. Earlier, his 1993 paper on describing and segmenting scenes from incomplete data (13 citations) addressed the fundamental challenge of visual perception under uncertainty. Nevatia also led the development of the DRIVE (Dynamic Reasoning from Integrated Visual Evidence) framework for the DARPA Strategic Computing program’s autonomous land vehicle effort, as detailed in his 1992 work (5 citations). His research has profoundly influenced autonomous systems, object detection, and video analysis, inspiring generations of computer vision researchers.