Yuanqing Lin

Papers

Total Citations

H-Index

About

Yuanqing Lin is a researcher whose work bridges robotics, computer vision, and signal processing, with a particular focus on perception and localization. His key research areas include appearance-based robot localization, nonlinear manifold learning, and acoustic source localization. In his influential 2005 paper on "Learning nonlinear appearance manifolds for robot localization," Lin proposed a novel method for estimating a robot's low-dimensional pose from high-dimensional panoramic images by assuming the images lie on a nonlinear appearance manifold embedded in a high-dimensional space. This work, with 27 citations, demonstrated how local geometric structure could be exploited for robust visual localization. Earlier, in 2004, Lin contributed to acoustic signal processing with his work on "Nonnegative deconvolution for time of arrival estimation," also garnering 27 citations. Here, he generalized interaural time difference (ITD) estimation—a primary cue for sound source localization—by relating traditional cross-correlation methods to maximum-likelihood estimation and introducing a nonnegative deconvolution framework. These contributions showcase Lin's ability to apply advanced mathematical techniques to practical perception problems, making his work valuable for students and researchers interested in robot navigation, manifold learning, and multi-modal sensing.