Jing Zeng

Papers

Total Citations

H-Index

About

Jing Zeng is a robotics and computer vision researcher whose work sits at the intersection of autonomous exploration, 3D scene reconstruction, and simultaneous localization and mapping (SLAM). His research focuses on enabling robots to intelligently perceive and reconstruct complex environments with minimal human intervention, leveraging cutting-edge representations such as implicit neural networks and 3D Gaussian splatting. Zeng's most recognized contribution, "NeurAR: Neural Uncertainty for Autonomous 3D Reconstruction With Implicit Neural Representations" (2023), has garnered 71 citations and represents a significant advance in active reconstruction — addressing the challenging problem of how a robot can plan its own viewing path to build accurate scene models in real time. This work bridges the gap between offline neural rendering quality and the demands of live robotic operation. His more recent work extends this paradigm to multi-robot systems with semantic guidance, pushing the frontier toward scalable, collaborative scene understanding. His survey on Visual SLAM further reflects a commitment to synthesizing and grounding research progress for the broader community. Collectively, Zeng's contributions are shaping how autonomous agents perceive, map, and interact with the physical world — making his work highly relevant for researchers in robotics, embodied AI, and 3D vision.