Max Robotics — Global Service Robot Directory + US Certification

About

Duy-Quang Vu is a rising researcher at the intersection of computer vision and robotic manipulation, whose work focuses on enabling robots to perceive and interact with their environments more intelligently. His core research areas include object pose estimation, grasp detection, and hand-object interaction modeling, with a particular emphasis on making these systems robust to real-world clutter and uncertainty. Vu’s most influential work, “Graspability-Aware Object Pose Estimation in Cluttered Scenes” (18 citations), introduces a novel approach that considers whether an object can actually be grasped when estimating its pose—a critical insight for practical robotics. He has also made significant contributions to grasp synthesis from 3D point clouds using attention mechanisms (17 citations) and to multi-modal hand-object pose estimation with adaptive fusion (11 citations). Notably, Vu addresses a key limitation in the field by developing methods for collision-free grasp detection from color and depth images (9 citations) and attention-based grasp detection using monocular depth estimation (5 citations), reducing reliance on expensive depth sensors. His work on attention-based hand pose estimation with voting and dual modalities (8 citations) further demonstrates his commitment to advancing perception systems for augmented reality, virtual reality, and imitation-based robot learning.