Shuai Liao

I obtained my Ph.D. from University of Amsterdam, the Netherlands, focusing on computer vision and deep learning.

I did internship at Qualcomm AI Research, Netherlands in 2020. I was a visiting research intern at the University of Queensland, Australia in 2014.

Download Resume  

profile photo
Research Interests

I am interested in 3D computer vision and video understanding. I have a good understanding of multiple view geometry and I am passionate about Autonomous driving and AR/VR. In particular, I have experience with:

  1. 3D computer vision:
         Viewpoint estimation;    6DOF pose estimation;    3D object detection;
         3D rotation estimation;    Surface normal estimation;    Depth estimation.

  2. Image classification:
         Multi-label classification;    Zero-shot learning;    Adversarial attacks and defenses;    Credible classification/confidence calibration.

  3. Video understanding:
         Optical flow estimation;    Video object segmentation/detection.

News

[Sep. 22, 2021]      I have recently defended my doctorate of philosophy. 🎉



Publications          [Full list]
b3do

Quasibinary Classifier for Images with Zero and Multiple Labels
Shuai Liao, Efstratios Gavves, ChangYong Oh, Cees G. M. Snoek
ICPR, 2020
poster / bibtex

It is well-known that one-vs.-rest classification and multi-label classification need to be handled by the right classifier, i.e. softmax classifier vs. sigmoid classifiers. Could it be simpler? Yes. Quasibinary classifier might be the only classifier you need. And even better, Quasibinary classifier yields well-calibrated probabilities that are critical to tasks such as autonomous driving and clinical diagnosis.





Spherical Regression: Learning Viewpoints, Surface Normals and 3D Rotations on n-Spheres
Shuai Liao, Efstratios Gavves, Cees G. M. Snoek
CVPR, 2019
project page / bibtex / arXiv / supplemental materials

Regression problems have always been tricky for deep neural networks to learn, mainly due to unstable gradient. To resolve this, we proposal spherical regression, a plugin module that boosts the performance for three regression tasks in computer vision at the same time. We also present a 3D rotation estimation dataset called ModelNet10-SO3.

Searching and Matching Texture-free 3D Shapes in Images
Shuai Liao, Efstratios Gavves, Cees G. M. Snoek
ICMR, 2018
poster / bibtex

We present a pipeline that automatically selects and aligns textureless 3D shapes (e.g. CAD models) onto 2D images. This is particular useful for collecting training data for autonomous driving.

b3do Tag Features for Geo-Aware Image Classification
Shuai Liao, Xirong Li, Heng Tao Shen Yang Yang Xiaoyong Du
IEEE Transactions on Multimedia (TMM), 2015
bibtex
b3do Zero-shot Image Tagging by Hierarchical Semantic Embedding
Xirong Li, Shuai Liao, Weiyu Lan, Xiaoyong Du, Gang Yang
SIGIR, 2015
project page / bibtex
b3do Building geo-aware tag features for image classification
Shuai Liao, Xirong Li, Xiaoxu Wang, Xiaoyong Du
IEEE International Conference on Multimedia and Expo (ICME), 2014
bibtex
Patents

Learning geometric differentials for matching 3D models to objects in a 2D image
Shuai Liao, Efstratios Gavves, Cornelis Gerardus Maria Snoek
(US Patent 2020)

Coordinate estimation on n-spheres with spherical regression
Shuai Liao, Efstratios Gavves, Cornelis Gerardus Maria Snoek
(US Patent 2020)



Academic Service
Reviewer for

Conference on Neural Information Processing Systems (NeurIPS)

International Conference on Learning Representations (ICLR)

Conference on Computer Vision and Pattern Recognition (CVPR)

European Conference on Computer Vision (ECCV)

AAAI Conference on Artificial Intelligence (AAAI)

Transactions on Multimedia Computing Communications and Applications(TOMM)

ACM Multimedia (ACMMM)

ACM International Conference on Multimedia Retrieval (ICMR)

Teaching for

Computer Vision Master course, 2016

Applied Machine Learning Master course 2018, 2019

Behavior-Based Robotics Bachelor course, 2018

Supervision for

Master thesis: Pose estimation of ships at port of Rotterdam. (2019-2020)








Website adapted from this source code.