Papers
3
Total Citations
32
H-Index
2
About
Kunlong Hong is a leading researcher at the intersection of robotics, computer vision, and infrastructure inspection, with a primary focus on automating the detection and assessment of surface defects in critical civil structures. His major contributions lie in developing integrated robotic systems that combine deep learning, 3D reconstruction, and neural radiance fields to inspect dam spillways—a task traditionally requiring dangerous manual labor. Hong’s work on “Multiple Defects Inspection of Dam Spillway Surface Using Deep Learning and 3D Reconstruction Techniques” (15 citations) introduces a robotic solution that automates the detection of concrete deterioration after prolonged scouring. He further advanced the field with “Inspection-NeRF” (15 citations), which synthesizes multi-type local images—original, semantic, and depth—from a global 3D model, enabling operators to inspect specific defect locations in detail. Additionally, his research on dynamic SLAM (2 citations) enhances robot localization accuracy in environments with moving objects by fusing semantic information with geometric constraints. Hong’s innovative fusion of climbing robots, neural rendering, and defect detection is setting new standards for safe, efficient, and high-fidelity infrastructure inspection.
Research Focus
Key Achievements
Top Papers
- 1
- 2
- 3