About

I am an ECE Ph.D. candidate advised by Patricio A. Vela at Institute for Robotics and Intelligent Machines, Georgia Tech. I also collaborate closely with Jonathan Tremblay, Stephen Tyree, Bowen Wen, and Stan Birchfield from NVIDIA Research. I am a practical roboticist, interested in combining robotics and computer vision together to solve the challenges in the real world. To be more specific, I am working on 6-DoF Object Pose Estimation, Object Grasping, and Human-robot Interaction related topics. [View my CV] (Updated in Jan. 2023).

I received B.E. in Automation from Southeast University in 2018, advised by Wenze Shao and Yangang Wang. I was also a research intern at Applied Nonlinear Control Lab, University of Alberta, in 2017, advised by Alan Lynch; NVIDIA Learning and Perception Research group from May 2020 to May 2021 and May 2022 to Dec 2022, advised by Stan Birchfield.

News

Interests

  • Robotics
  • Computer Vision
  • Image Processing

Education

  • Ph.D. in Electrical and Computer Engineering, 2023 (Expected)

    Georgia Institute of Technology

  • M.S. in Electrical and Computer Engineering, 2020

    Georgia Institute of Technology

  • B.E. in Automation, 2018

    Southeast University

Featured Publications

Parallel Inversion of Neural Radiance Fields for Robust Pose Estimation

A parallelized optimization method based on fast Neural Radiance Fields (NeRF) for estimating 6-DoF target poses.

Yunzhi Lin, Thomas Müller, Jonathan Tremblay, Bowen Wen, Stephen Tyree, Alex Evans, Patricio A. Vela, Stan Birchfield
ICRA 2023
Preprint Project

Keypoint-Based Category-Level Object Pose Tracking from an RGB Sequence with Uncertainty Estimation

A single-stage, category-level 6-DoF pose estimation algorithm that simultaneously detects and tracks instances of objects within a known category.

Yunzhi Lin, Jonathan Tremblay, Stephen Tyree, Patricio A. Vela, Stan Birchfield
ICRA 2022
Preprint Code Project

Single-Stage Keypoint-based Category-level Object Pose Estimation from an RGB Image

A single-stage, keypoint-based approach for category-level object pose estimation that operates on unknown object instances within a known category using a single RGB image as input.

Yunzhi Lin, Jonathan Tremblay, Stephen Tyree, Patricio A. Vela, Stan Birchfield
ICRA 2022
Preprint Code Project

Multi-view Fusion for Multi-level Robotic Scene Understanding

A system for multi-level scene awareness for robotic manipulation, including three types of information: 1) a point cloud representation of all the surfaces in the scene, for the purpose of obstacle avoidance. 2) the rough pose of unknown objects from categories corresponding to primitive shapes (e.g., cuboids and cylinders), and 3) full 6-DoF pose of known objects.

Yunzhi Lin, Jonathan Tremblay, Stephen Tyree, Patricio A. Vela, Stan Birchfield
IROS 2021
Preprint PDF Dataset Project Video

Using Synthetic Data and Deep Networks to Recognize Primitive Shapes for Object Grasping

A segmentation-based architecture proposed to decompose objects into multiple primitive shapes from monocular depth input for robotic manipulation.

Yunzhi Lin, Chao Tang, Fujen Chu, Patricio A. Vela
ICRA 2020
Preprint PDF Code Project Video

Publications

Quickly discover relevant content by filtering publications.
Parallel Inversion of Neural Radiance Fields for Robust Pose Estimation
Yunzhi Lin, Thomas Müller, Jonathan Tremblay, Bowen Wen, Stephen Tyree, Alex Evans, Patricio A. Vela, Stan Birchfield
IEEE International Conference on Robotics and Automation (ICRA 2023)
Preprint Project
Keypoint-Based Category-Level Object Pose Tracking from an RGB Sequence with Uncertainty Estimation
Yunzhi Lin, Jonathan Tremblay, Stephen Tyree, Patricio A. Vela, Stan Birchfield
IEEE International Conference on Robotics and Automation (ICRA 2022)
Preprint Code Project
SGL: Symbolic Goal Learning in a Hybrid, Modular Framework for Human Instruction Following
Ruinian Xu, Hongyi Chen, Yunzhi Lin, Patricio A. Vela
IEEE Robotics and Automation Letters
Preprint
See all publications

Contact