Xingtian Qu

Papers

Total Citations

H-Index

About

Xingtian Qu is a researcher whose work centers on computer vision and 3D scene understanding, with a particular focus on monocular depth estimation for autonomous systems. His most cited paper, "Superb Monocular Depth Estimation Based on Transfer Learning and Surface Normal Guidance" (2020, 8 citations), introduces a novel approach that leverages a lightweight Convolutional Neural Network (CNN) for coarse depth prediction, enhanced by surface normal guidance and transfer learning. This contribution addresses a critical challenge for drones and robots—accurately perceiving 3D environments for path planning and navigation—without relying on heavy computational resources. Qu’s method stands out for its efficiency and accuracy, making it practical for real-time applications in robotics and autonomous navigation. By integrating transfer learning, he demonstrates how pre-trained models can be adapted to depth estimation tasks, reducing the need for extensive labeled datasets. With 8 citations, this work has already influenced peers in the field, showcasing Qu’s ability to bridge theoretical advances with deployable solutions. His research continues to push the boundaries of how machines interpret spatial data, offering promising pathways for smarter, more autonomous systems.