Leisheng Zhong

I am currently a Ph.D. candidate at the Department of Electronic Engineering in Tsinghua University, Beijing, China.

I received the B.Eng. degree in Electronic Engineering from Tsinghua University in 2013.

I have been pursuing my Ph.D. degree since 2015 in Institute of Information Cognition & Intelligent System, Tsinghua University under the supervision of Prof. Li Zhang.

I worked as an intern at Tencent Autonomous Driving Center during the summer of 2019.

Email  /  CV  /  中文主页

Research

My research interests include 3D computer vision, robotic vision, augmented reality, computer graphics and machine learning. I am currently mainly working on robust 3D object pose estimation and tracking.

TIP2020 Occlusion-aware Region-based 3D Pose Tracking of Objects with Temporally Consistent Polar-based Local Partitioning
Leisheng Zhong, Xiaolin Zhao, Yu Zhang, Shunli Zhang, Li Zhang
IEEE Transactions on Image Processing (TIP), 2020
preprint / bibtex
IJCV2019 A Robust Monocular 3D Object Tracking Method Combining Statistical and Photometric Constraints
Leisheng Zhong, Li Zhang
International Journal of Computer Vision (IJCV), 2019
preprint / bibtex
TCSVT2018 A Direct 3D Object Tracking Method Based on Dynamic Textured Model Rendering and Extended Dense Feature Fields
Leisheng Zhong, Ming Lu, Li Zhang
IEEE Transactions on Circuits and Systems for Video Technology (TCSVT), 2018
preprint / poster / bibtex
Projects and Demos
lidar

A Stereo-Lidar SLAM System

We build a Simultaneously Localization and Mapping (SLAM) system based on a line-scan Lidar and two cameras.

poster / video: youtube youku

AR_demo

3D Object Tracking for Augmented Reality

Some Augmented Reality (AR) demos based on our research about 3D object tracking.

video: youtube youku

3dpaint

3D Paint

Our 3D object pose tracking algorithm makes it possible to paint on the surface of 3D models, instead of on 2D papers.

video: youtube youku

edge_tracking_demo

Fast 3D Object Tracking with Edge Distance Fields

Very fast occlusion-aware 6-DOF object pose tracking based on edge distance fields. Objects could be tracked with only ~100 edge points.

video: youtube youku

uarm

A Vision-aided Robot Arm

We assemble a webcam to a commercial robot arm (uArm swift pro) and develop some demo applications, including automatic pick-and-place, laser engraving, 3D printing, planar target tracking, and the simulation of air refueling.

video: youtube youku

Honors and Awards

First-Class Academic Scholarship in Tsinghua, 2018

National Scholarship for Doctoral Students, 2017

Outstanding Graduates of Beijing, 2013

National Scholarship for Undergraduate Students, 2012

Tsinghua-Samsung Scholarship, 2011



The source code of this website is adapted from this page.