Xu Zhang

Applied Scientist

About Me

I’m an applied scientist at Amazon Alexa. I was a postdoctoral research scientist at Digital Video and MultiMedia Lab at the Columbia University in the city of New York, supervised by Prof. Shih-Fu Chang. I obtained a Ph.D. from Tsinghua University, Beijing, China, in 2016, supervised by Prof. Shengjin Wang. My research interests include local feature detection and description, large scale image retrieval and image forensics. Here is my full resume.


Jun. 2020

I have one paper accepted by IEEE Journal of Selected Topics in Signal Processing

May. 2020

I'll be serving as a Special Session Chair in WIFS 2020

Feb. 2020

I'll be serving as an AC in ACMMM 2020

Sep. 2019

I have one paper accepted in WIFS 2019


Amazon Alexa

Cambridge, MA, USA

Applied Scientist

Nov. 2019 - Present

Columbia University

New York, New York, USA

Postdoctoral Research Scientist

July 2016 - Oct. 2019

Tencent, Inc.

Beijing, China

Research Intern

Nov. 2015 - May 2016


Beijing, China

Research Intern

Aug. 2015 - Nov. 2015


Tsinghua University

Beijing, China

Ph.D., Electrical Engineering

Sep. 2013 - Jul. 2016

Columbia University

New York, New York, USA

Visiting Scholar

Nov. 2014 - May 2015

Tsinghua University

Beijing, China

M.S., Electrical Engineering

Sep. 2010 - Jul. 2013

Tsinghua University

Beijing, China

B.S., Electronic Engineering

Sep. 2006 - Jul. 2010


Visual Political Literacy Project

Given a protest image, detect and recognize all political symbols in the image. The project helps journalists recognized different symbols. The project has been covered by [The Economist] [CNET] [project page]


The goal of this software is to reconstruct semantically meaningful 3D models of buildings and other man-made structures from satellite imagery. I'm responsible for the building segmentation and building roof type segmentation tasks. [project page]


NYC Media Lab’s Annual Summit 2019

Brooklyn, NY

Sep. 2019

We will host a demo "Political Visual Literacy App" [website]

CVPR 2019

Long Beach, CA

Jun. 2019

I co-organized the 2nd Landmark Recognition Workshop [website]

CVPR 2018

Salt Lake City, UT

Jun. 2018

I co-organized the 1st Landmark Recognition Workshop [website]

Selected Publication

Xu Zhang, Zhaohui H. Sun, Svebor Karaman, Shih-Fu Chang

Discovering Image Manipulation History byPairwise Relation and Forensics Tools [paper] [bibtex] [code]

IEEE Journal of Selected Topics in Signal Processing

Xu Zhang, Svebor Karaman, Shih-Fu Chang

Detecting and Simulating Artifacts in GAN Fake Images [paper] [bibtex] [code]

WIFS 2019

Mang Ye, Xu Zhang, Pong C. Yuen, and Shih-Fu Chang

Unsupervised Embedding Learning via Invariant and Spreading Instance Feature [paper] [bibtex] [code]

CVPR 2019

Matthew J. Leotta, Chengjiang Long, Bastien Jacquet, Matthieu Zins, Dan Lipsa, Jie Shan, Bo Xu, Zhixin Li, Xu Zhang, Shih-Fu Chang, Matthew Purri, Jia Xue, Kristin Dana

Urban Semantic 3D Reconstruction From Multiview Satellite Imagery [paper] [bibtex] [code]

CVPRW 2019. Best paper in the EarthVision 2019 Workshop

Xu Zhang, Felix X Yu, Sanjiv Kumar, Shih-Fu Chang

Learning Spread-out Local Feature Descriptors [paper] [bibtex] [code]

ICCV 2017. Spotlight

Xu Zhang, Felix X Yu, Svebor Karaman, Shih-Fu Chang

Learning Discriminative and Transformation Covariant Local Feature Detectors [paper] [bibtex] [code]

CVPR 2017

Xu Zhang, Felix X Yu, Ruiqi Guo, Sanjiv Kumar, Shengjin Wang, Shih-Fu Chang

Fast Orthogonal Projection Based on Kronecker Product [paper] [bibtex] [code]

ICCV 2015

Xu Zhang, Felix Xinnan Yu, Svebor Karaman, Wei Zhang, Shih-Fu Chang

Heated-Up Softmax Embedding [paper] [bibtex] [code]

arXiv 2018

Xu Zhang, Felix X Yu, Shih-Fu Chang, Shengjin Wang

Deep Transfer Network, Unsupervised Domain Adaptation [paper] [bibtex] [code]

arXiv 2015

A Little More About Me

I’m a big fan of scuba diving, photography and volleyball.