Junqiu Wang

Peking University, Osaka Research Institute of Industrial Science and Technology, The University of Osaka

Papers

Total Citations

H-Index

About

Junqiu Wang’s research lies at the intersection of computer vision, robotics, and intelligent sensing, with a core focus on enabling machines to perceive, localize, and interact with unstructured environments. His most influential work, "Image-based Localization and Pose Recovery Using Scale Invariant Features" (2005), pioneered the use of local scale-invariant features as natural landmarks for mobile robot localization, demonstrating remarkable robustness to occlusion and outliers—a foundational contribution to vision-based navigation. Wang further advanced object tracking and segmentation through two key papers (2010), where he introduced novel frameworks that integrate appearance and spatial information of patches, as well as min-cut optimization with consecutive shape priors, to handle cluttered and dynamic backgrounds. These methods directly address challenges in human-computer interaction, video surveillance, and autonomous robotics. While his citation counts (9, 3, and 2) reflect a focused, early-career impact, his work is notable for its practical, algorithmic elegance—offering computationally efficient solutions to persistent problems in real-world vision systems. Wang’s contributions remain relevant for researchers developing robust, feature-based localization and segmentation pipelines in robotics and intelligent sensing.