Ziyi Wu (吴紫屹)

  Ph.D. Student

  University of Toronto

  Email: ziyiwu [at] cs.toronto.edu

  CVGoogle ScholarGitHub


I am a first-year Ph.D. student in People, AI and Robotics (PAIR) group at the University of Toronto, supervised by Prof. Animesh Garg and Prof. Andrea Tagliasacchi. I am affiliated with the Vector Institute. I also work closely with Dr. Thomas Kipf.

Prior to my Ph.D., I received my Bachelor's degree from the Department of Automation, Tsinghua University, where I was fortunate to work with Prof. Jiwen Lu. I spent an unforgettable summer in 2020 doing internship at Stanford University, advised by Prof. Leonidas J. Guibas.

My research interests are in computer vision and robotics, particularly 3D scene understanding and their applications to robotics. I am broadly interested in designing machine learning systems that can perceive the world as we human do, which I believe is the cornerstone of achieving AGI.




SlotFormer: Long-Term Dynamic Modeling in Object-Centric Models
Ziyi Wu, Nikita Dvornik, Klaus Greff, Jiaqi Xi, Thomas Kipf*, Animesh Garg*
UAI@CRL workshop, 2022. (* equal supervision)


Breaking Bad: A Dataset for Geometric Fracture and Reassembly
Silvia Sellán*, Yun-Chun Chen*, Ziyi Wu*, Animesh Garg, Alec Jacobson
Under review. (* equal contribution)
[Project Page


Dynamics-aware Adversarial Attack of 3D Sparse Convolution Network
An Tao, Yueqi Duan, He Wang, Ziyi Wu, Pengliang Ji, Haowen Sun, Jie Zhou, Jiwen Lu
Under review.
[Paper]  [Code


IF-Defense: 3D Adversarial Point Cloud Defense via Implicit Function based Restoration
Ziyi Wu*, Yueqi Duan*, He Wang, Qingnan Fan, Leonidas J. Guibas
Preprint. (* equal contribution)
[Paper]  [Code]  [Slide]


Instance Similarity Learning for Unsupervised Feature Representation
Ziwei Wang, Yunsong Wang, Ziyi Wu, Jiwen Lu, Jie Zhou
ICCV, 2021.
[Paper]  [Code]


Learning Efficient Binarized Object Detectors with Information Compression
Ziwei Wang, Jiwen Lu, Ziyi Wu, Jie Zhou
T-PAMI, 2021.


BiDet: An Efficient Binarized Object Detector
Ziwei Wang, Ziyi Wu, Jiwen Lu, Jie Zhou
CVPR, 2020.
[Paper]  [Code]  [Slide]


A Cascade Regression Model for Anatomical Landmark Detection
Zimeng Tan, Yongjie Duan, Ziyi Wu, Jianjiang Feng, Jie Zhou
MICCAI Workshop, 2019.
[Paper]  [Slide]


CFUN: Combining Faster R-CNN and U-net Network for Efficient Whole Heart Segmentation
Zhanwei Xu, Ziyi Wu, Jianjiang Feng, Jie Zhou
[Paper]  [Code]  [Slide]

Academic Services

Journal reviewer: ISRR'22; Conference reviewer: CVPR'22; ECCV'22; NeurIPS'22; IROS'22;
Workshop reviewer: OSC@ICLR'22;


Teaching Assistant, CSC 108: Introduction to Computer Programming21 Fall, 21-22 Winter

Selected Awards


I like swimming and I am particularly good at freestyle. I also like basketball. My favorite player is LeBron James.

I am also an Anime fan and a lover of Visual Novel. Some of my favorite works are Fate/Stay Night, Hyouka, Steins;Gate, EVA and Subarashiki Hibi (a.k.a. Wonderful Everyday).

Two persons I admire the most: Kaiming He for his outstanding research tastes and Elon Musk for his aspiration and enthusiasm.

I write an article summarizing the experience during my CS PhD application process. Take a look here (written in Chinese)!

Ziyi Wu Last updated: June.13, 2022