[Conference News] Likelihood Measurement Model Helps in Face Recognition

A wide range of potential applications in law-enforcement and security has made face recognition an important part of research in computer vision and pattern recognition. However, low-resolution surveillance videos with uncontrolled pose and illumination present a significant challenge to face tracking and recognition algorithms. In “Face recognition in Low-Resolution Videos Using Learning-Based Measurement Model,” [] presented at the 2011 International Joint Conference on Biometrics (IJCB 2011), researchers from the University of Notre Dame propose using a learning-based likelihood measurement model to handle the large appearance and resolution difference between gallery images and probe videos. The measurement model consists of a map that transforms gallery and probe features into a space where their inter-Euclidean distances approximate the distances that would have been obtained had all the descriptors been computed from good-quality frontal images.

Papers from IJCB 2011 are available to both IEEE Computer Society members and paid subscribers via the Computer Society Digital Library.