Hamed Hosseini

Papers

Total Citations

H-Index

About

Hamed Hosseini is a roboticist whose research focuses on the intersection of computer vision and robotic manipulation, specifically in the domain of intelligent grasp detection. His work aims to solve the fundamental challenge of enabling robots to perceive and interact with objects in unstructured environments with human-like dexterity. Hosseini’s key contributions include developing an improved pipeline for real-time robotic grasp detection using Convolutional Neural Networks, which can identify optimal grasp rectangles for both seen and unseen objects—a paper that has garnered 16 citations. He further advanced the field with his AGILE framework, which teaches robot manipulators to infer optimal grasps by learning from the approach angle of the gripper, mimicking human hand-object positioning. Most recently, his work on multi-modal robust geometry primitive shape abstraction proposes a novel method for simplifying complex scenes into predefined geometric forms, combining simulation and real-world data to enhance grasp success in unknown environments. Through these contributions, Hosseini is pushing the boundaries of autonomous robotic manipulation, making strides toward more adaptable and intelligent robotic systems capable of operating beyond controlled factory settings.