Khanh-Duong Tran

Papers

Total Citations

H-Index

About

Khanh-Duong Tran is a rising researcher at the forefront of robotic manipulation and human-computer interaction, with a core focus on computer vision and deep learning. His work bridges the critical gap between perception and action, specifically targeting collision-free grasp detection and precise hand pose estimation. In his highly cited 2024 paper, "Collision-Free Grasp Detection From Color and Depth Images" (9 citations), Tran addresses a fundamental bottleneck in robotics: generating reliable grasp poses from multimodal data. He innovatively combines color and depth information to overcome the limitations of point cloud data, which lacks appearance cues, thereby enabling more robust and efficient robotic grasping. Complementing this, his work "Efficient Multimodal Fusion for Hand Pose Estimation With Hourglass Network" (6 citations) tackles the complex, real-time challenge of tracking highly articulated hand movements for applications in VR, AR, and gesture recognition. By fusing different data modalities within a powerful hourglass network architecture, Tran achieves a balance of speed and accuracy essential for interactive systems. With a growing citation impact, his contributions are laying the groundwork for more dexterous robots and seamless human-machine interfaces.