Huini Fu

National University of Defense Technology

Papers

Total Citations

H-Index

About

Huini Fu is a researcher whose work lies at the intersection of computer vision and robotics, with a particular focus on depth estimation and perceptual learning. Her most-cited paper, "Unsupervised Stereo Depth Estimation Refined by Perceptual Loss" (2018), addresses a critical challenge in mobile robotics: accurately inferring object depth from binocular images without relying on expensive ground-truth labels. By integrating perceptual loss into an unsupervised learning framework, Fu’s approach refines depth maps to achieve greater fidelity, bridging the gap between traditional methods and supervised deep learning successes. This contribution has garnered attention in the field, with 3 citations, and highlights her ability to innovate in data-efficient vision systems. Fu’s work is especially relevant for autonomous navigation and 3D scene understanding, where robust depth estimation is essential. Her research demonstrates a commitment to advancing practical, scalable solutions in computer vision, making her a notable voice among emerging scholars in this domain.