Max Robotics — Global Service Robot Directory + US Certification

About

Van-Hiep Duong is a rising researcher at the forefront of computer vision and robotic manipulation, with a focused expertise in hand and object pose estimation. His work directly addresses the core challenge of enabling robots to perceive and interact with their environment with human-like dexterity. Duong’s major contributions lie in developing novel, attention-based, and multimodal fusion architectures that significantly improve the accuracy and robustness of pose estimation in cluttered, real-world scenes. He has pioneered techniques for "graspability-aware" object pose estimation, ensuring that a robot’s perceived object orientation is not just accurate but also actionable for a successful grasp. His research on multi-modal hand-object pose estimation, which adaptively fuses RGB and depth data, is critical for advancing applications in augmented reality, virtual reality, and imitation-based robot learning. With his most-cited paper, "Graspability-Aware Object Pose Estimation in Cluttered Scenes," already garnering 18 citations within its first year, Duong’s work is rapidly gaining recognition for its practical impact. His consistent output of high-quality, attention-based models for both hand and grasp detection marks him as a key innovator bridging the gap between perception and robotic action.