Research

 Face Processing

In this age of videos and surveillance cameras, the human face is without a doubt the most important object in image analysis.  Our group is interested in many aspects of the human face: modeling, rendering, animating, identity recognition, expression analysis, and even psychological theories of human perception.  Our work finds applications in surveillance, security, entertainment, education.  The following papers provide a snapshot of what we do.

• Hamed Kiani and Terence Sim. "Correlation Filter Cascade for Facial Landmark Localization." 
  IEEE Winter Conference on Applications of Computer Vision, 2016. [pdf]

• Galoogahi, Hamed Kiani, Terence Sim, and Simon Lucey. "Multi-Channel Correlation Filters."
  In Computer Vision (ICCV), 2013 IEEE International Conference on, pp. 3072-3079. IEEE, 2013. [pdf]

• Kiani Galoogahi, Hamed, and Terence Sim. "Face photo retrieval by sketch example."
  In Proceedings of the 20th ACM international conference on Multimedia, pp. 949-952. ACM, 2012. [pdf]

• ZHANG*, L, N YE, E. M. Martinez and T Sim, "Expressive deformation profiles for
  cross-expression face recognition". International Conference on Pattern Recognition (2012). [pdf]

• HOSSEIN*, N, T Sim and E. M. Martinez, "Do you see what I see? A more realistic eyewitness
  sketch recognition". IEEE Biometrics Compendium (2011). New York: IEEE. (International
  Joint Conference on Biometrics, 11 - 13 Oct 2011. [pdf]

• SONG*, Z, B NI, D GUO, S YAN and T Sim, "Learning universal multi-view age estimator by
  video contexts". International Conference on Computer Vision (2011). [pdf]

• GUO*, D and T Sim, "Face Makeup by Example". Proceedings of the IEEE Conference on
  Computer Vision and Pattern Recognition (2009). [pdf]

• Sim*, T, S Zhang, J LI and Y CHEN, "Simultaneous and orthogonal decomposition of data
  using mulitmodal discirminant analysis". IEEE International Conference on Computer Vision
  (2009). [pdf]

• Zhang*, S and T Sim, "Discriminant subspace analysis: a Fukunaga-Koontz approach". IEEE
  TRANSACTIONS ON PATTERN ANALYSIS AND MACHINE INTELLIGENCE, 29, no. 10
  (2007): 1732 - 1745. [pdf]

• Zhang*, Y, T Sim and C L Tan, "Realistic and efficient wrinkle simulation using an
  anatomy-based face model with adaptive refinement". Computer Graphics International 2005
  (2005) [pdf]

• Sim*, T, S Baker and M Bsat, "The CMU Pose, Illumination, and Expression Database". IEEE
  TRANSACTIONS ON PATTERN ANALYSIS AND MACHINE INTELLIGENCE, 25, no. 12
  (2003): 400-420. [pdf] (This work has over 1500 citations, according to Google Scholar.)

Biometrics

Going beyond face recognition, our work in biometrics explores other ways to identify people, including identical twins. We also pioneered the concept of Continuous Authentication using Biometrics, that is, allowing a computer system to continuously determine whether the authorized user is always the one using the device.  This is to prevent hijacking, where an imposter forcibly takes over a system after the legitimate user has been authorized.  Selected publications below.

• HOSSEIN*, N, L ZHANG, T Sim, E. M. Martinez and D GUO, "Wonder ears: identification of
  identical twins from ear images". International Conference on Pattern Recognition 2012. [pdf]

• ZHANG*, L, N YE, E. M. Martinez, D GUO and T Sim*, "New hope for recognizing twins by
  using facial motion". IEEE Workshop on Applications of Computer Vision (2012). [pdf]

• YE*, N and T Sim, "Towards general motion-based face recognition". IEEE Conference on
  Computer Vision and Pattern Recognition (2010). [pdf] [video]

• Sim*, T, S Zhang, R Janakiraman and S Kumar, "Continuous verification using multimodal
  biometrics". IEEE TRANSACTIONS ON PATTERN ANALYSIS AND MACHINE
  INTELLIGENCE, 29, no. 4 (Biometrics) (2007): 687 - 700. [pdf]

• JANAKIRAMAN*, R and T Sim, "Keystroke dynamics in a general setting". Advances in
  Biometrics (2007): 584 - 593. Berlin: Springer Verlag. International Conference on
  Biometrics, 27 - 29 Aug 2007. [pdf]

Computational Photography

Digital photography is the best thing since sliced bread!  Really. You can do so much more with a digital camera than with a film camera.  Face detection is only a start. Imagine a camera that produces picture-perfect photographs every time: images without blur, without red-eyes, that faithfully capture the ambience at your candlelight dinner, that reveal far more detail in bright daylight.  Only a dream? Not according to our group of researchers!

• Shaojie Zhuo, Terence Sim. "Defocus map estimation from a single image"
  Pattern Recognition, Volume 44, Issue 9, September 2011, Pages 1852-1858. [pdf]

• GUO*, D, Y CHENG, S ZHUO and T Sim, "Correcting over-exposure in photographs". IEEE
  Conference on Computer Vision and Pattern Recognition (2010). [pdf]

• YE*, N, T Sim and X.P. Miao, "Video stylization by single image example". IEEE International
  Conference on Image Processing (2010). [pdf] [video]

• ZHUO*, S, X ZHANG, XP Miao and T Sim, "Enhancing low light images using near infrared
  flash images". IEEE International Conference on Image Processing (2010). [pdf]

• GUO*, D and T Sim, "Color me right: seamleass image compositing". Lecture notes in
  computer science 5702 (2009). Berlin: Springer Verlag. (International Conference on
  Computer Analysis of Images and Patterns, 2 - 4 Sep 2009) [pdf]

• ZHANG*, X, Miao, X and T Sim, "Enhancing photographs using near infrared images". IEEE Computer
  Society Conference on Computer Vision and Pattern Recognition (2008). [pdf]