Max Robotics — Global Service Robot Directory + US Certification

About

Lingni Ma is a leading researcher in 3D computer vision and robotics, whose work bridges the gap between photorealistic scene reconstruction and intelligent semantic understanding. Her most impactful contribution is the **Replica dataset** (2019, 384 citations), a landmark resource of 18 high-fidelity, semantically annotated indoor 3D scenes that has become a standard benchmark for embodied AI and scene understanding research. Ma’s core expertise lies in developing deep learning methods for consistent semantic mapping from RGB-D cameras (2017, 148+ citations), where she pioneered multi-view neural networks that enforce spatial and temporal coherence across frames—a critical capability for robots navigating dynamic environments. She has also advanced practical 3D reconstruction by introducing plane priors to de-noise, stabilize, and complete online reconstructions in real time, enabling robust performance on mobile devices for AR/VR and robotics. Her earlier work on planar simplification and texturing of dense point clouds (2013) addressed the scalability challenge of processing multi-million point datasets. With over 600 total citations, Ma’s research continues to shape how machines perceive and interact with complex indoor spaces, making her work essential reading for anyone building the next generation of spatially aware autonomous systems.