Graphics enhanced version of this site

Face Recognition Across Pose
Head: Simon Lucey
Contact: Simon Lucey (slucey+@andrew.cmu.edu)

Mailing address:
Carnegie Mellon University
Robotics Institute
5000 Forbes Avenue
Pittsburgh, PA 15213

Associated center: VASC
Associated labs/groups: Human Identification at a Distance, Face Group, and Human Sensing

In many face recognition tasks the pose of the probe and gallery images are different. In other cases multiple gallery or probe images may be available, each captured from a different pose. We have developed a face recognition algorithm that is able to handle large variations in pose using a "patch-based" rather than "holistic" representations. The algorithm operates by estimating the "eigen light-field" of the subject's head from the input gallery or probe images. Matching between the probe and gallery is then performed using the eigen light-fields.

Our algorithm drammatically out-performs, FaceIT, a commercially available system from Visionics Corporations. A brief summary of our results on the PIE Database are:

Name - Title <Email Address>
• [Home] Ralph Gross - PhD Student, ISRI <rgross@cs.cmu.edu>
• [Home] Simon Lucey - Systems Scientist <slucey+@andrew.cmu.edu>
• [Home] Iain Matthews - Senior Systems Scientist <iainm@cs.cmu.edu (Currently On Leave)>

• Learning Patch Correspondences for Improved Viewpoint Invariant Face Recognition
  A.B. Ashraf, S. Lucey, and T. Chen
  IEEE International Conference on Computer Vision and Pattern Recognition (CVPR), June, 2008. [Abstract]
  Download: pdf [948 KB] copyrighted

• A Viewpoint Invariant, Sparsely Registered, Patch Based, Face Verifier
  S. Lucey and T. Chen
  International Journal of Computer Vision (IJCV), December, 2007. [Abstract]
  Download: pdf [734 KB] copyrighted

• Integrating Monolithic and Free-parts Representations for Improved Face Verification in the Presence of Pose Mismatch
  S. Lucey and T. Chen
  Pattern Recognition Letters, Vol. 28, No. 8, June, 2007, pp. 895 - 903. [Abstract]

• A Viewpoint Invariant, Sparsely Registered, Patch Based, Face Verifier
  S. Lucey and T. Chen
  tech. report CMU-RI-TR-07-17, Robotics Institute, Carnegie Mellon University, February, 2007. [Abstract]
  Download: pdf [1554 KB] copyrighted

• Learning patch dependencies for improved pose mismatched face verification
  S. Lucey and T. Chen
  IEEE International Conference on Computer Vision and Pattern Recognition (CVPR), June, 2006. [Abstract]
  Download: pdf [538 KB] copyrighted

• Using overlapping distributions to deal with face pose mismatch
  S. Lucey
  British Machine Vision Conference (BMVC), September, 2005. [Abstract]
  Download: pdf [184 KB] copyrighted

• Face Recognition Across Pose and Illumination
  R. Gross, S. Baker, I. Matthews, and T. Kanade
  Handbook of Face Recognition, Stan Z. Li and Anil K. Jain, ed., Springer-Verlag, June, 2004. [Abstract]
  Download: pdf [617 KB] copyrighted

• Appearance-Based Face Recognition and Light-Fields
  R. Gross, I. Matthews, and S. Baker
  IEEE Transactions on Pattern Analysis and Machine Intelligence, Vol. 26, No. 4, April, 2004, pp. 449 - 465. [Abstract]
  Download: pdf [775 KB] copyrighted

• Fisher Light-Fields for Face Recognition Across Pose and Illumination
  R. Gross, I. Matthews, and S. Baker
  Proceedings of the German Symposium on Pattern Recognition (DAGM), September, 2002. [Abstract]
  Download: pdf [258 KB], ps.gz [3085 KB] copyrighted

• Appearance Based Face Recognition and Light-Fields
  R. Gross, I. Matthews, and S. Baker
  tech. report CMU-RI-TR-02-20, Robotics Institute, Carnegie Mellon University, August, 2002. [Abstract]
  Download: pdf [5701 KB] copyrighted