Yuliang Zou

Ph.D. student
Department of Electrical and Computer Engineering
Virginia Tech
ylzou [at] vt [dot] edu

About

Bio

I am a first-year Ph.D. student in Computer Engineering at Virginia Tech. I work with Prof. Jia-Bin Huang. My recent research interests focus on deep generative modeling and video analysis.

Before that, I have worked with Prof. Honglak Lee, Dr. Yuting Zhang and Ruben Villegas during my masters at University of Michigan, Ann Arbor, and Prof. Xiaojin Gong during my undergrad at Zhejiang University.

In 2017 summer, I had my wonderful time visiting California and working with Zelun Luo and Dr. Judy Hoffman.

My CV is available here (last updated on Nov. 2017).

Education

Aug. 2017 - Present

Aug. 2015 - Apr. 2017

Sep. 2011 - Jun. 2015

Industry

May. 2018 - Aug. 2018

Teaching

Teaching Assistant     

ECE 5554 / ECE 4554 (Computer Vision), Fall 2017 [link] [tutorial] [Guest Lecture on Object Detection]

News

[2018/07]     

Our paper on Unsupervised Joint Learning of Depth and Flow has been accepted to ECCV. See you in Munich!

[2018/07]     

Our paper on Human-Object Interaction Detection has been accepted to BMVC.

[2018/05]     

The code of our ICML2017 paper is released!

[2018/04]     

I passed my Ph.D. Qualifying Examination!

[2018/04]     

I will be an official reviewer for NVIDIA AI City Challenge Workshop.

[2018/02]     

I will intern at Adobe, advised by Dr. Jimei Yang (Summer 2018).

Publications

DF-Net: Unsupervised Joint Learning of Depth and Flow using Cross-Network Consistency

Depth Estimation Optical Flow Unsupervised Learning

Yuliang Zou, Zelun Luo, Jia-Bin Huang

In Proceedings of the 29th European Conference on Computer Vision (ECCV), 2018.

[arXiv] [Code] [Project Page]

iCAN: Instance-Centric Attention Network for Human-Object Interaction Detection

Object Detection Scene Understanding

Chen Gao, Yuliang Zou, Jia-Bin Huang

In Proceedings of the 29th British Machine Vision Conference (BMVC), 2018

[arXiv] [Code] [Project Page]

Label-Efficient Learning of Transferable Representations across Domains and Tasks

Domain Adaptation Semi-Supervised Learning Transfer Learning

Zelun Luo, Yuliang Zou, Judy Hoffman, Li Fei-Fei

In Proceedings of the 31st Neural Information Processing Systems (NIPS), 2017

[arXiv] [Project Page]

Learning to Generate Long-term Future via Hierarchical Prediction

Generative Modeling Video Analysis

Ruben Villegas, Jimei Yang, Yuliang Zou, Sungryull Sohn, Xunyu Lin, Honglak Lee

In Proceedings of the 34th International Conference on Machine Learning (ICML), 2017

[arXiv] [Code] [Poster] [Project Page] [Slides] [SOHU News (Chinese)]

Projects

2016

Automatic Group Photograph Enhancement

Face Detection Image Manipulation

Yuliang Zou, Zhusheng Wang, Yayan Zhao, Enpei Xi

[Report] [Code]

Single-Machine MapReduce Server

Distributed System

Yuliang Zou

[Code]

Free Food Calendar

Information Retrieval Web Development

Yuliang Zou*, Haoran Wang*, Sang Hyuk Cho*, Connor Larson*

[Code]

VR-Chat: Virtual Reality Based Online Chatroom

Kinect Mobile Development Unity3D Virtual Reality

Peifeng Yu, Rui Chen, Yuliang Zou, Yixiang Wu

[Code] [Website]

Misc

Relevant Courses Taken

ECE5454 - Optimization Techniques

CS5984 - Deep Learning

EECS442 - Computer Vision

EECS498 - Information Retrieval

EECS551 - Matrix Methods

EECS542 - Advanced Topics in Computer Vision

EECS545 - Machine Learning

EECS598 - Random Matrix Theory

CSE - Numerical Method

ISEE - Fundamentals of Information Theory

ISEE - Fundamentals of Digital Video and Audio

cs294a - Unsupervised Feature Learning and Deep Learning

cs231n - Convolutional Neural Networks for Visual Recognition

ee364a - Convex Optimization I

UCL_RL - Reinforcement Learning (Ongoing)

Reading Group

I am taking charge of the VT Vision and Learning Reading Group this semester. If you are at VT, feel free to join us.

About My Name

My given name is Yuliang (瑜亮), which comes from two famous characters in the Three Kindoms. And my family name is Zou (邹).

My English name is Zack. You can also call me Y.L. Anyway, call whatever you like. :)

Contact