Max Robotics — Global Service Robot Directory + US Certification

About

Lintao Zheng is a researcher specializing in 3D computer vision, autonomous robotics, and deep learning for shape understanding. His work sits at the intersection of active perception, scene reconstruction, and object recognition, with a particular focus on enabling robots to intelligently explore and interpret unknown environments. Zheng's most influential contribution is VERAM (2018), a view-enhanced recurrent attention model for 3D shape classification that introduced a principled view selection mechanism into multi-view deep learning, garnering 81 citations and advancing the field of active object recognition. His earlier work on attention-driven depth acquisition (2016, 49 citations) tackled the challenge of fine-grained shape identification through autonomous, sequential depth sensing — a problem with direct implications for real-world robotics applications. His research on tensor-field-guided autonomous reconstruction (2017, 47 citations) demonstrated how robot navigation paths could be optimized to balance exploration efficiency with reconstruction quality, a theme he has continued to refine through semantic scene understanding and hand-eye calibration. Collectively, his body of work, accumulating over 230 citations, reflects a sustained commitment to making robotic perception smarter, more purposeful, and more adaptive in complex real-world settings.