Ngoc-Trung Ho
Papers
6
Total Citations
60
H-Index
5
About
Ngoc-Trung Ho is a rising researcher at the intersection of computer vision and robotics, whose work focuses on enabling robots to perceive and interact with their environments more intelligently. His primary research areas include object pose estimation, hand-object interaction modeling, and robotic grasp synthesis. Ho’s major contributions lie in developing attention-based and multi-modal fusion techniques that allow robots to accurately estimate the pose of objects and hands in cluttered, real-world scenes—a critical capability for autonomous manipulation. His most cited work, “Graspability-Aware Object Pose Estimation in Cluttered Scenes” (2024, 18 citations), introduces a novel approach that considers whether an object can actually be grasped, moving beyond simple geometric pose recovery. He has also advanced grasp detection by using monocular depth estimation to bypass the need for expensive 3D sensors, as seen in his 2024 paper on the topic. With over 60 cumulative citations across his top papers, Ho is establishing himself as a key voice in making robotic manipulation more robust and accessible, with applications spanning augmented reality, virtual reality, and imitation-based robot learning.
Research Focus
Key Achievements
Top Papers
- 1Graspability-Aware Object Pose Estimation in Cluttered Scenes18 citations · 2024
- 2Grasp Configuration Synthesis from 3D Point Clouds with Attention Mechanism17 citations · 2023
- 3
- 4Attention-based hand pose estimation with voting and dual modalities8 citations · 2024
- 5Attention-Based Grasp Detection With Monocular Depth Estimation5 citations · 2024
- 6Vote-based multimodal fusion for hand-held object pose estimation1 citations · 2025