Guangming Zhu
Papers
7
Total Citations
434
H-Index
7
About
Guangming Zhu is a leading researcher in human-robot interaction and intelligent sensing, with a focus on gesture and action recognition using multimodal data. His work bridges computer vision and robotics, particularly through the use of RGB-D sensors like Kinect. Zhu’s most cited paper, “Multimodal Gesture Recognition Using 3-D Convolution and Convolutional LSTM” (277 citations), introduces a novel deep learning framework that fuses spatial and temporal features for robust gesture recognition, setting a benchmark in the field. He also pioneered online continuous human action recognition (57 citations), enabling real-time human-robot collaboration, and developed algorithms for fast robot identification and mapping (23 citations) and whole-body motion imitation for humanoid robots (22 citations). Beyond robotics, Zhu has explored semantic scene completion from depth images (33 citations) and even contributed to materials science with photodirected 2D-to-3D morphing structures (10 citations). His work is widely cited for its practical impact on autonomous systems and intelligent interfaces, making him a key figure in advancing how machines perceive and interact with human motion.
Research Focus
Key Achievements
Top Papers
- 1Multimodal Gesture Recognition Using 3-D Convolution and Convolutional LSTM277 citations · 2017
- 2
- 3Semantic scene completion with dense CRF from a single depth image33 citations · 2018
- 4A Fast Robot Identification and Mapping Algorithm Based on Kinect Sensor23 citations · 2015
- 5Fast human whole body motion imitation algorithm for humanoid robots22 citations · 2016
- 6Human action recognition using key poses and atomic motions12 citations · 2015
- 7