Lei Zhang


Google Scholar     Github     Linkedin    
Dexterous Robotic Manipulation; Computer Vision; Deep Learning; Reinforcement Learning

Education:
University of Hamburg (UHH, Germany) | Ph.D. Student
Leibniz Universität Hannover (LUH, Germany) | M.Sc., Sep. 2017 - Mar. 2020
Harbin Institute of Technology (HIT, China) | B.Sc., Sep. 2012 - June. 2016

FFHClutteredGrasping: Multi-fingered Robotic Hand Grasping in Cluttered Environments through Hand-object Contact Semantic Mapping

Under Review
Keywords: Generation Model; Multi-fingered Robotic Hand; Grasping from Clutter Scenes.
Webpage  •   PDF  •   Demo  •   FFH Cluttered Grasping Dataset

ToolEENet: Tool Affordance 6D Pose Estimation

Under Review
We present the innovative TOOLEE dataset, which, to the best of our knowledge, is the first to feature affordance segmentation of a tool's end-effector (EE) along with its defined 6D pose based on its usage and we propose the ToolEENet framework for accurate 6D pose estimation of the tool's EE. Our approach excels in this field, demonstrating high levels of precision and generalization. Furthermore, it shows great promise for application in contact-based manipulation scenarios. Keywords: 6D Pose Estimation, Multi-fingered Robotic Hand, Tool Use.
Webpage  •   PDF  •   Code  •   Dataset

CG-CNN: A Collision-Aware Cable Grasping Method in Cluttered Environment

Accepted at ICRA2024
We introduce a Cable Grasping-Convolutional Neural Network (CG-CNN) designed to facilitate robust cable grasping in cluttered environments. Given our model’s implicit collision sensitivity, we achieved commendable success rates of 92.3% for known cables and 88.4% for unknown cables, surpassing contemporary state-of-the-art approaches. Keywords: Collision Awareness; Grasping from Clutter Scenes with Complicated Objects; Sim-to-Real Tranfer.
Webpage  •   PDF  •   Demo  •   Cable Grasping Dataset

Close the Sim2real Gap via Physically-based Structured Light Synthetic Data Simulation

Accepted at ICRA2024
We proposed a novel method to simulate structured light camera and generated realistic dataset for object detection, segmentation, robotic grasping tasks.
Keywords: Structured-light Simulation for Generation of Realistic Data; Data Generation for Perception; Reduce Sim2Real Gap.
Webpage  •   PDF  •   Demo  •   Photorealistic Structured-light Dataset

Towards Precise Model-free Robotic Grasping with Sim-to-Real Transfer Learning

Lei Zhang, Kaixin Bai, Zhaopeng Chen, Yunlei Shi, Jianwei Zhang
2022 IEEE International Conference on Robotics and Biomimetics (IEEE ROBIO 2022).
★ Best Conference Paper Award Finalist, ROBIO ★
Keywords: Data augmentation for Generation of Dense Grasping Labels; Sim-to-Real Transfer; Model-free Grasping Dataset.
Webpage  •   PDF  •   Demo

Learning of 6D Object Poses with Multi-task Point-wise Regression Deep Networks

Kaixin Bai, Lei Zhang, Zhaopeng Chen, Jianwei Zhang
2022 IEEE International Conference on Robotics and Biomimetics (IEEE ROBIO 2022).
Keywords: Deep Learning-based 6D Pose Estimation; 6D Rotation Representation; 6D Pose Estimation-based Robotic Grasping.
WebPage  •  

Maximizing the Use of Environmental Constraints: A Pushing-Based Hybrid Position/Force Assembly Skill for Contact-Rich Tasks

Yunlei Shi, Zhaopeng Chen, Lin Cong, Yansong Wu, Martin Craiu-Müller, Chengjie Yuan, Chunyang Chang, Lei Zhang, Jianwei Zhang
2022 IEEE International Conference on Robotics and Biomimetics (IEEE ROBIO 2021).
Keywords: Contact-Rich Assembly; Hybrid Position/Force.
PDF  •