Cheng-Kai Lu
Papers
2
Total Citations
12
H-Index
1
About
Cheng-Kai Lu is a researcher at the forefront of computer vision and human-robot interaction, with a focus on monocular tracking, mapping, and multimodal emotion recognition. His work bridges model-based and data-driven approaches, as highlighted in his highly cited 2022 review on monocular tracking and mapping—a comprehensive survey that has garnered 11 citations for its systematic comparison of traditional geometric methods versus modern deep learning techniques. This contribution has become a valuable resource for researchers navigating the evolution of visual SLAM and 3D reconstruction. More recently, Lu has advanced companion robotics with a lightweight multimodal deep learning framework for real-time emotion recognition, integrating facial and speech features. Designed for resource-constrained platforms like the Zenbo Junior II robot, this 2025 study employs a customized GhostNet with Triplet Attention Modules and a Frame Attention Network, achieving efficient, real-time performance. By enabling robots to perceive human emotions with minimal computational overhead, Lu’s work directly enhances natural human-robot interaction, making assistive technologies more responsive and empathetic. His research consistently pushes the boundaries of efficient, intelligent systems, offering practical solutions for robotics and embedded vision.
Research Focus
Key Achievements
Top Papers
- 1
- 2