Yidong Li

Papers

Total Citations

H-Index

About

Yidong Li is a researcher whose work spans computer vision, robotics, and deep learning, with a particular focus on enhancing perception and autonomy in challenging environments. His most cited paper, "Pedestrian Detection with Super-Resolution Reconstruction for Low-Quality Image" (2021, 63 citations), addresses a critical real-world problem: improving object detection accuracy when input images are degraded. By integrating super-resolution techniques, Li’s work enables more reliable pedestrian detection in low-resolution surveillance or autonomous driving scenarios, directly impacting safety and system robustness. In robotics, Li has contributed to planar object tracking through his work on "Constrained Confidence Matching" (2018), proposing a novel algorithm that overcomes drift and fast-motion failures common in conventional template trackers—a key advancement for robotic manipulation and visual servoing. He has also provided valuable surveys, such as "Attention Models for Point Clouds in Deep Learning" (2021), synthesizing the state of the art in 3D data representation for computer vision and robotics. Additionally, his research on "Joint Optimization of Path Planning and Task Assignment for Space Robot" (2021) tackles the complex, resource-constrained problem of coordinating multiple space robots, demonstrating his versatility in applying optimization to aerospace applications. With a growing citation record and contributions that bridge theory and practical deployment, Li’s work is shaping more capable, resilient autonomous systems.