Renshu Gu

Papers

Total Citations

191

H-Index

About

Renshu Gu is a computer vision researcher whose work centers on multi-object tracking (MOT), a critical challenge in applications ranging from autonomous driving and traffic flow analysis to robotic vision and surveillance systems. Gu is best known for developing TrackletNet, an innovative framework that addresses one of the field's most persistent problems: maintaining reliable target tracking in the face of unreliable detection, occlusion, and rapid camera motion. By exploiting the connectivity between tracklets, Gu's approach offers a more robust solution to target re-identification and trajectory linking under challenging real-world conditions. The 2019 iteration of this work has accumulated 167 citations, demonstrating significant influence within the computer vision community, while the earlier 2018 conference version further established the foundational ideas behind the framework. Gu's research sits at the intersection of deep learning and practical visual perception, contributing tools that are directly applicable to safety-critical systems such as self-driving vehicles. For students and researchers working on detection-based tracking pipelines, Gu's TrackletNet work remains a meaningful reference point for graph-based and connectivity-aware tracking strategies.