Konstantinos G. Derpanis

Toronto Metropolitan University, University of Pennsylvania, York University

Papers

Total Citations

279

H-Index

About

Konstantinos G. Derpanis is a leading researcher in computer vision and robotics, with a focus on 3D object detection, pose estimation, and semantic scene understanding. His most influential work, "Single image 3D object detection and pose estimation for grasping" (2014, 227 citations), introduced a pioneering deformable parts-based model that enables robots to detect objects and estimate their 3D pose from a single image, even when objects lack texture cues. This breakthrough directly supports robotic grasping in cluttered environments, bridging vision and manipulation. Derpanis also advanced robot localization in "Robot localization using soft object detection" (2012, 39 citations), where he reimagined the problem by using object class detection—rather than traditional visual features—to achieve data association with semantically annotated or hand-drawn maps. This work opened new avenues for intuitive, human-like navigation. Earlier in his career, he contributed to teleconferencing with the SAVI system (2001, 13 citations), an actively controlled setup that enhanced remote collaboration. With a citation count exceeding 279 across his top works, Derpanis’s research has had a lasting impact on autonomous systems, particularly in enabling machines to perceive and interact with the physical world through robust, semantically rich vision.