No abstract

We describe the design and development of a prototype whole-hand imaging system. The sensor is based on multispectral technology that is able to provide hand shape, fingerprints and palmprint modalities of a user's hand by a single user interaction with the sensor. A clear advantage of our system over other unimodal sensors for these modalities include: (i) faster acquisition time, (ii) better quality images, and (iii) ability to provide spoof detection. Initial results on a medium-size database show good recognition performance based on individual modalities as well as after fusing multiple fingers and fusing finger and palm. The prototype is being refined in order to improve performance even further.

No abstract

This paper presents the first investigation into the classification of faces from unconstrained video sequences in natural scenes, i.e., with arbitrary poses, facial expressions, occlusions, illumination conditions and motion blur. To overcome difficulties from individual frames, a novel Bayesian formulation is proposed to estimate the posterior probability of a face trait at a specific time, conditional on features identified in previous frames of a video sequence. A Markov model is used to represent temporal dependencies, and classification involves determining the maximum a posteriori class at a given time. Showing the robustness of the proposed system, the Bayesian framework is first trained on a database collected under controlled conditions, and then applied to the previously unseen faces obtained from an unconstrained video database. The Markovian temporal model results in a gender classification rate of 90% by the last video frame, and is shown to outperform alternative approaches previously introduced in the literature.