Changyin Zhou

Read as: /chan in/        /zou/

 

 
  Ph.D Candidate      
  Computer Vision Lab, Computer Science Department Office: 616 CEPSR    
  Columbia University in the City of New York Telephone: 1-212-939-7090    
  New York, NY 10027 Email: Firstname.Lastname AT gmail.com    
           
 

Hello, my name is Changyin Zhou. I have graduated from the CAVE lab of Computer Science Department, Columbia U. with a Ph.D degree. My advisor at Columbia U. was Prof. Shree Nayar. My Ph.D. research was mainly focused on computational imaging and physics-based vision, and my Ph.D. thesis is titled Point-spread-function Engineering for Scene Recovery.

I received my MS in Computer Science and BS in Statistics, both from Fudan University, China. My master thesis was on pattern classification and feature selection. I worked for Microsoft Global Technical Engineering Center as an engineer from 2001 to 2003. I worked with Dr. Stephen Lin on image enhancement in Microsoft Research Asia in 2006, with Dr. Brian Guenter on camera and lens design in Microsoft Research Redmond in the summer of 2010, and with Dr. Kari Pulli on mobile computational photography in Nvidia Research in the summer of 2011.

 
           

 

I am currently a software engineer of Google [x].

 
           

 

     
  Publications  
         
 
  Focal Sweep Photography for Space-time Refocusing (new!)

Changyin Zhou, Daniel Miau, and Shree Nayar

Space-time refocus: www.focalsweep.com

[project page][pdf (coming soon)][©]

 

 
         
         
 
  Robust Stereo with Flash and No-flash Image Pairs (new!)

Changyin Zhou, Alejandro Troccoli, and Kari Pulli

CVPR 2012 (poster)

[pdf][poster][©]

 

 
         
    Aperture Evaluation for Defocus Deblurring and Extended Depth of Field (new!)

Changyin Zhou and Shree Nayar

Technical Report, CUCS-007-12, Columbia University, April 2012

[pdf] [©]

(A short version appears in ICCP'09.)

 

 
         
 
  Computational Cameras: Convergence of Optics and Processing

Changyin Zhou and Shree Nayar

IEEE Transactions on Image Processing (TIP), Volume 20, Issue 12, Pages 3322-3340, 2011

[preprint] [©] [errata]

See my candidacy talk for a basic idea of this review paper.

 
         
         
 

  Coded Aperture Pairs for Depth from Defocus and Defocus Deblurring

Changyin Zhou, Stephen Lin, and Shree Nayar

International Journal of Computer Vision (IJCV), Volume 93, Number 1, Pages 53-72, 2011

[pdf] [©]

This is an extended version of [Zhou et al., ICCV 2009].

 
         
         
    Programmable Aperture Camera Using LCoS

Hajime Nagahara, Changyin Zhou, Takuya Watanabe, Hiroshi Ishiguro, and Shree Nayar

ECCV 2010 [pdf] [©]

The real-time LCoS implementation for DFD and light field capturing won Japan MIRU2011 demo award.

 
         
         
    Diffusion Coded Photography for Extended Depth of Field

Oliver Cossairt, Changyin Zhou, and Shree Nayar

SIGGRAPH 2010

[project page] [pdf] [supplementary images] [derivations] [©]

Patent with S.K. Nayar and O. Cossairt, Jan 2011

 
         
 
    Depth from Diffusion

Changyin Zhou, Oliver Cossairt, and Shree Nayar

CVPR 2010 (oral presentation)

[project page] [pdf] [©]

 
         
         
    Coded Aperture Pairs for Depth from Defocus

Changyin Zhou, Stephen Lin, and Shree Nayar

ICCV 2009 (oral presentation)

[project page] [pdf] [supplementary materials] [code] [©]

 
         
         
    What are Good Apertures for Defocus Deblurring?

Changyin Zhou and Shree Nayar

ICCP 2009 (oral presentation)

[project page] [pdf] [slides] [code and data][©]

 
         
         
    Flexible Depth of Field Photography

Hajime Nagahara, Sujit Kuthirummal, Changyin Zhou, and Shree Nayar

ECCV 2008 (oral presentation)

[project page] [pdf] [©]

Patent with S.K. Nayar, H. Nagahara, and S. Kuthirummal, Oct 2009

 
 
 
    Removal of Image Artifacts Due to Sensor Dust

Changyin Zhou and Stephen Lin

CVPR 2007 (oral presentation)

[pdf] [slides] [©]

Feature: http://research.microsoft.com/en-us/news/features/cvpr2007.aspx

Patent with B. Guo and S. Lin, Dec 2007

 
         
         
    Improving Nearest Neighbor Classification with Cam Weighted Distance

Changyin Zhou and Yanqiu Chen

Pattern Recognition, Volume 39, Issue 4, Pages 635-645, 2006

[pdf] [©]

 
         
         
 

* In computer vision, two leading journals are the International Journal of Computer Vision (IJCV), and the IEEE Transactions on Pattern Analysis and Machine Intelligence (PAMI). Leading conferences are IEEE CVPR, ICCV and ECCV. Papers in these conferences are rigorously reviewed, with acceptance rates around 20%, and below 5% for oral presentations.

 

 

     
  Course Projects  
         
    Full Body 3-D Scanning

with Sam Calabrese and Abhishek Gandhi

[project page][report][slides] [©]

"This project is to create/manuplate a 3D model of a live human."

Course: 3D Photography, Fall 2007, Prof. Peter Allen

 
       

 

 
    Virtual Computational Camera

Changyin Zhou

[project page] [report] [slides] [©]

"This project is to formulate/implement computational camera at an abstract level."

Course: Computational Photography, Spring 2009, Prof. Peter Belhumeur

 
       

 

 
    Color Sparse Coding

Changyin Zhou

[project page] [report] [slides]

"This project is to extract/manipulate 'basic' image colors from a single image."

Course: Advanced Machine Learning, Fall 2008, Prof. Tony Jebara

 
       

 

 
    Human-aided Classification

Changyin Zhou and Bingrui Yang

[Human Classifier (exe)][slides]

"This project is to evaluate/leverage the classification power of human being."

Course: Biometrics, Fall 2007, Prof. Peter Belhumeur

 
         
         

 

   
  Talks
       
 
  • ICCP, Apr 2009, San Francisco: What are good apertures for defocus deblurring? [slides] [©]
 
  • ICCV, Oct 2009, Kyoto, Japan: Coded aperture pairs for depth from defocus. [slides] [©]
 
  • CVPR, Jun 2010, San Francisco: Depth from diffusion. [slides] [©]
 
  • SIGGRAPH, Jul 2010, Los Angels: Diffusion Coded Photography for Extended Depth of Field. [slides] [©]
 
  • Microsoft Research Summer Internship Report, July 2010, Redmond: Designing the Ultimate Camera.
 
  • Ph.D Candidacy Talk, Jan 2011: Computational Cameras for Light Field Processing. [slides] [paper list] [©]

  

   
  Teaching Assistant
       
 
  • COMS W4731: Computer Vision, Fall 2008 and Spring 2010
 
  • COMS E6732: Computational Imaging, Spring 2008
 
  • Design and Analysis of Algorithms, CSE, Fudan University, Fall 2005

 

   
  Honors and Certificates
       
 
  • Microsoft Research Asia Best Intern (10 out of 400+), 2007 (News feature in Chinese)

  • Honor of Excellent Master Graduate Student, Shanghai, 2007

  • First-prize People's Scholarship (Dong's Dongfang Scholarship), Fudan Univ., 2005

  • Microsoft Certified Solution Developer (MCSD) 2001

  • Microsoft Certified Systems Engineer (MCSE), 2001

  • Microsoft Certified Database Administrator (MCDBA), 2001

  • Senior Software Engineer, certified by Chinese Ministry of Information Industry, 1998

   

 

     
  Services  
         
 
  • Reviewer / program committee: Pattern Recognition, Computing Reviews, PAMI, IJCV, TIP, Signal Processing, ICCV 2009, Siggraph Asia 2009, ACCV 2010, Siggraph Asia 2010, Siggraph Asia 2011, ICCV 2011, ICCP 2011, CHI 2011, CVPR 2012, ECCV 2012
 
     

 

   
  Resources
       
 
  • Good to Read - A paper list on computer vision related topics