Xiuming Zhang 张修明

I am a Ph.D. candidate advised by Prof. William T. Freeman in the Computer Science and Artificial Intelligence Laboratory (CSAIL) at Massachusetts Institute of Technology (MIT). I work in the fields of computer vision, computational photography, and computer graphics, with particular interests in relighting, view synthesis, and material modeling.

 xiuming at csail dot mit dot edu  |  GitHub  |  Google Scholar

Past Experiences


Apr. 2019MoSculp is on exhibition in the MIT Museum (photos available here)
Jan. 2019Awarded Snap Research Fellowship
Dec. 2018We presented GenRe at NeurIPS and released our codebase for training/testing GenRe and ShapeHD
Sep. 2018MoSculp got featured as the 9/19 MIT homepage and received media coverage by BBC, Forbes, ACM, MIT, and etc.
Oct. 2016Our paper on Bayesian modeling of Alzheimer's disease heterogeneity is out in PNAS and received media coverage by EurekAlert!, MGH/HMS, NUS, and etc.
May 2016Our work on Alzheimer's disease heterogeneity received the Magna Cum Laude Award and oral presentation at ISMRM

Recent Publications

* indicates equal contribution.

Press Coverage

Side Work

Non-Referred Notes and Side Projects


My Python library that includes useful functions for signal/image processing, 3D modeling and rendering with Blender, visualization utility, and etc.
Documentation  |  Code

ReML Estimation in Linear Models

A tutorial on Restricted Maximum Likelihood Estimation in linear regression and Linear Mixed-Effects Models.


A MATLAB implementation excluding border matting and user editing.

Basic Photometric Stereo

A MATLAB implementation of the basic photometric stereo algorithm.

Bandgap of 2D Photonic Crystal

Replicating a classic textbook plot with COMSOL-MATLAB scripting.


Conference Reviewer

ECCV 2020, CVPR 2020, AAAI 2020, ICCV 2019, ICML 2019, SIGGRAPH 2019, MM 2019, GLOBECOM 2018

Journal Reviewer