Yawei Du

Papers

Total Citations

H-Index

About

Yawei Du is a researcher advancing the field of robotic manipulation through deep learning, with a particular focus on grasp detection for parallel-plate grippers. Their work centers on developing end-to-end convolutional neural network architectures that enable robots to predict single or multiple grasping poses from RGB and depth images. Du’s major contributions include the introduction of the Inverted Residual Convolutional Neural Network (IR-ConvNet) model, which leverages inverted residual blocks to improve efficiency and accuracy in grasp detection. This work, detailed in their 2022 paper, demonstrates a modular robotic system capable of robust, real-time grasp planning. Du further extended this research with their 2023 study on cooperative grasp detection, exploring multi-grasp scenarios to enhance robotic dexterity. While their citation counts are still growing—with 2 citations each for their key papers—these works represent foundational steps in applying lightweight neural networks to practical robotics. Du’s research is particularly relevant for students and engineers interested in computer vision, deep learning, and autonomous manipulation, offering a clear pathway from model design to real-world robotic application.