I am a computer scientist in Adobe Systems by the beautiful Lake Union in Seattle, Washington. I survived my Ph.D. program in the Image Processing Lab, Dept. of Electrical & Computer Engineering, University of Minnesota. I am interested in computer vision, machine learning, image/video editing, real-time interactive systems, etc.
Email:  xubai AT adobe DOT com (work)

 xbai0721 AT gmail DOT com (personal)

Tel: 206-675-7291 (work)
Office: 801 N 34th St, Seattle, WA 98103, USA


Ph.D. 2010, Electrical & Computer Engineering, University of Minnesota, Twin Cities.

Adviser: Prof. Guillermo Sapiro

M.S. 2007, Electrical & Computer Engineering, University of Minnesota, Twin Cities.

B.S. 2005, Electrical Engineering, Shanghai Jiao Tong University, Shanghai, China.


Geodesic Segmentation and Matting 

A novel framework for interactive image/video segmentation and matting. The algorithm runs in O(N) time.

Accepted as Oral Presentation in ICCV 2007.


Video SnapCut

A robust video object cutout system designed for high-quality video production.


Video SnapCut, renamed Roto Brush, was transferred to Adobe After Effects CS5 as a major new feature on April 12th, 2010.

More links: Adobe TV, Creative Cow Library, Macworld, YouTube, AETuts, ... (Google after effects roto brush)


Dynamic Color Flow


Temporally Coherent Video Matting


Refine Edge for After Effects CC (New)

Finally a practical tool for video matting!!!

Tutorials: Adobe TV, Pro Video Coalition, Mamoworld, History of Rotoscoping in AE, ...




● Xue Bai, Guillermo Sapiro, Distancecut: interactive segmentation and matting of images and videos. IEEE ICIP 2007 (International Conference on Image Processing), San Antonio, Texas.
● Xue Bai, Guillermo Sapiro, A geodesic framework for fast interactive image and video segmentation and matting. IEEE ICCV 2007 (International Conference on Computer Vision), Rio De Janeiro, Brazil.  (oral) 
● Xue Bai, Guillermo Sapiro, Geodesic Matting: A Framework for Fast Interactive Image and Video Segmentation and Matting. International Journal on Computer Vision 2008. 
● Xue Bai, Vladimir Cherkassky, Gender Classification of Human Faces Using Inference through Contradictions. 2008 International Joint Conference on Neural Networks.  

Xue Bai, Jue Wang, David Simons, Guillermo Sapiro, Video SnapCut: Robust Video Object Cutout Using  Localized Classifiers. ACM Transaction on Graphics (Proc. SIGGRAPH 2009). 

Xue Bai, Jue Wang, Guillermo Sapiro, Dynamic Color Flow: A Motion-Adaptive Color Model for Object Segmentation in Video, ECCV 2010 (European Conference on Computer Vision), Crete, Greece. 

Xue Bai, Jue Wang, David Simons, Towards temporally-coherent video matting, International Conference on Computer Vision/Graphics Collaboration Techniques and applications, Mirage 2011, France

Courses & Projects

[Full List]



My Photographs



Free Web Counter


The views and opinions expressed in this page are strictly those of the page author.
The contents of this page have not been reviewed or approved by the University of Minnesota.