Mingkui Tan

Papers

Total Citations

259

H-Index

About

Mingkui Tan is a prominent researcher whose work spans **3D scene understanding, autonomous perception, and embodied AI navigation**. His most significant contributions lie at the intersection of multi-sensor fusion and semantic segmentation, where he has advanced how machines interpret complex 3D environments. His landmark paper, "Perception-Aware Multi-Sensor Fusion for 3D LiDAR Semantic Segmentation" (2021), has garnered nearly 200 citations, establishing him as a leading voice in the autonomous driving and robotics communities. This work, alongside its 2024 follow-up EPMF, demonstrated how intelligently combining RGB camera data with LiDAR point clouds can substantially improve scene understanding — a critical capability for safe autonomous vehicles. Beyond sensor fusion, Tan has made meaningful inroads into vision-and-language navigation and active robot perception. His research on weakly supervised map learning for navigating agents following natural language instructions reflects a broader ambition to build robots that reason about their environments more like humans do. His work on active camera control for multi-object navigation further underscores his commitment to practical, deployable robotic systems. Collectively, Tan's research portfolio represents a cohesive vision: enabling machines to perceive, interpret, and navigate the physical world with greater intelligence and efficiency.