Qingtian Zhu

Papers

Total Citations

H-Index

About

Qingtian Zhu is a computer vision researcher whose work centers on 3D reconstruction and multi-view stereo (MVS), with a particular focus on harnessing deep learning to advance the state of the art in scene understanding and spatial modeling. His most recognized contribution, the 2021 survey "Deep Learning for Multi-View Stereo via Plane Sweep," has garnered 17 citations and offers a comprehensive synthesis of how deep learning architectures have been adapted to solve the challenging plane sweep-based stereo problem — a foundational technique underpinning applications in autonomous driving, robotics, and virtual reality. Building on this, his 2024 follow-up survey on learning-based MVS methods broadens the scope further, cataloguing algorithmic advances in recovering dense 3D scene structures from multi-viewpoint imagery. Together, these works position Zhu as an emerging authority in survey scholarship within the 3D vision community, providing researchers and practitioners with structured roadmaps through a rapidly evolving field. His contributions are particularly valuable for students entering computer vision, offering clear, well-organized overviews of complex methodologies that bridge classical geometric approaches with modern neural network-driven techniques.