Zhidong Liang

Master in Robotics

Shanghai Jiao Tong University

Biography

NOTE: 中文（简体）网站请点击右上角球形图案

I am currently a perception algorithm engineer at Baidu Apollo researching on L4 perception and 3D point cloud detection.

I received my master degree from Shanghai Jiao Tong University (Inteliigent Vehicle Lab) in 2020, advised by Associate Professor Chunxiang Wang and Professor Ming Yang, and received my B.Eng from Shanghai Jiao Tong University in 2017.

My research interest focuses on 3D deep learning on point cloud and its applications on autonomous driving. In addition, I used to be the person in charge of the CyberTiggo self-driving system in the Intelligent Vehicle Lab, responsible for integrating and validating the self-driving system, and developing the decision-making, planning and control modules.

Interests

Autonomous Driving
3D Computer Vision
Point Cloud Segmentation/Detection

Education

Master in Mechanical Engineering, 2020

Shanghai Jiao Tong University, Inteliigent Vehicle Lab
Bachelor in Mechanical Engineering, 2017

Shanghai Jiao Tong University

Recent News

[2021.11] One paper accepted to IEEE Sensors Journal.

[2021.02] One paper accepted to CVPR 2021 Oral.

[2021.02] One paper accepted to IEEE Robotics and Automation Letters with joint ICRA 2021 option.

[2020.09] Our method RangeRCNN ranked 2nd on the KITTI 3D Car detection benchmark among all methods with papers.

[2020.07] Our method HRI-MSP-L ranked 1st on the KITTI cyclist detection benchmark for all four tasks (2D, 3D, BEV, Orientation).

[2020.06] One paper accepted to IEEE Robotics and Automation Letters with joint IROS 2020 option.

[2020.01] One paper accepted to Tsinghua Science and Technology.

[2019.02] Our method SALoss ranked first on the ScanNet benchmark (3D instance segmentation).

[2019.01] One paper accepted to ICRA 2019.

[2018.09] Our method HDGCN ranked first on the Paris-Lille-3D benchmark (3D semantic segmentation).

Publications

RangeLVDet: Boosting 3D Object Detection in LIDAR With Range Image and RGB Image

IEEE Sensors Journal

zehan zhang, zhidong liang, ming zhang, xian zhao, hao li, ming yang, wenming tan, shiliang pu

PDF

RangeIoUDet: Range Image based Real-Time 3D Object Detector Optimized by Intersection over Union

2021 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR 2021 Oral)

zhidong liang, zehan zhang, ming zhang, xian zhao, shiliang pu

PDF

CentroidReg: A Global-to-Local Framework for Partial Point Cloud Registration

IEEE Robotics and Automation Letters (also presented in ICRA 2021)

hengwang zhao, zhidong liang, chunxiang wang, ming yang

PDF

RangeRCNN: Towards Fast and Accurate 3D Object Detection With Range Image Representation

zhidong liang, ming zhang, zehan zhang, xian zhao, shiliang pu

PDF

3D Instance Embedding Learning With a Structure Aware Loss Function for Point Cloud Segmentation

IEEE Robotics and Automation Letters (also presented in IROS 2020)

zhidong liang, ming yang, hao li, chunxiang wang

PDF

3D Instance Embedding Learning With a Structure Aware Loss Function for Point Cloud Segmentation

See all publications