Minglei Lu

Papers

Total Citations

H-Index

About

Minglei Lu is a computer vision researcher whose work focuses on the intersection of 3D perception and scene understanding, particularly through stereo imaging. His most-cited paper, "Category-Level Object Detection, Pose Estimation and Reconstruction from Stereo Images" (2024), addresses a fundamental challenge in robotics and augmented reality: enabling machines to recognize, localize, and reconstruct novel objects within a known category without requiring per-instance CAD models. By leveraging stereo image pairs, Lu’s approach achieves robust 3D pose estimation and shape reconstruction, bridging the gap between 2D detection and full 3D understanding. This work has already garnered 5 citations since its publication, reflecting its timely relevance to the growing demand for generalizable object perception systems. Lu’s contributions are particularly significant for applications in autonomous navigation, warehouse automation, and AR/VR, where real-time, category-level understanding is critical. His research demonstrates a strong commitment to pushing the boundaries of geometric computer vision, and his early impact suggests a promising trajectory in advancing how machines interpret and interact with the physical world.