Max Robotics — Global Service Robot Directory + US Certification

About

Duc-Thanh Tran is a computer vision and robotics researcher whose work sits at the intersection of 3D perception, pose estimation, and robotic manipulation. His research focuses primarily on two interrelated domains: hand and object pose estimation, and grasp detection for autonomous robot systems — areas where accurate spatial understanding is essential for real-world deployment. Tran has made notable contributions to multimodal fusion techniques, developing architectures that intelligently combine RGB and depth data to overcome the individual limitations of each modality. His work on hand-object pose estimation employs adaptive fusion and interaction learning to tackle the notoriously difficult problem of recovering hand-object configurations during contact, with applications spanning augmented reality, virtual reality, and imitation learning for robots. On the manipulation side, his collision-free grasp detection and graspability-aware pose estimation frameworks advance robots' ability to reliably interact with cluttered, real-world environments. With a cumulative citation count approaching 60 across papers published solely in 2024–2025, Tran has rapidly established himself as a productive emerging voice in his field. His consistent attention to practical constraints — sensor availability, occlusion, and real-time performance — makes his research particularly relevant for students and engineers building deployable robotic systems.