No abstract

Biometrics is the technique of automatically recognizing individuals based on their biological or behavioral characteristics. Various biometric traits have been introduced and widely investigated, including fingerprint, iris, face, voice, palmprint, gait and so forth. Apart from identity, biometric data may convey various other personal information, covering affect, age, gender, race, accent, handedness, height, weight, etc. Among these, analysis of demographics (age, gender, and race) has received tremendous attention owing to its wide real-world applications, with significant efforts devoted and great progress achieved. This survey first presents biometric demographic analysis from the standpoint of human perception, then provides a comprehensive overview of state-of-the-art advances in automated estimation from both academia and industry. Despite these advances, a number of challenging issues continue to inhibit its full potential. We second discuss these open problems, and finally provide an outlook into the future of this very active field of research by sharing some promising opportunities.

Abstract-Face recognition is one of the most widely used biometric systems due to its non-intrusive, natural and easy to use characteristics. However, automatic face recognition becomes very challenging whenever the acquisition conditions are unconstrained. In order to enhance the robustness of face recognition under challenging conditions, this paper proposes to adopt a cohort-based score normalization procedure. Specifically, the polynomial regression-based cohort normalization is extended to the unconstrained face pair matching problem. Extensive experiments conducted on the LFW benchmark demonstrate the effectiveness of cohort normalization on this challenging scenario. Furthermore, we advance the state of the art in cohort normalization by providing a better understanding of its discriminative cohort behavior. In particular, we find that the cohort information alone has a certain discriminative power which is just marginally worse than the raw matching score. A larger cohort set size gives more stable and often better results to a point before the performance saturates and slightly reduces. Finally, the experimental results on both the FRGC ver 2.0 database (lab face verification) and the LFW database (wild face pair matching) show that cohort samples of different quality indeed produce different cohort normalization performance. Generally, for matching faces captured under lab environments, using cohort samples of good quality leads to much better performance than using bad cohort samples. However, for matching wild faces, using wild cohort achieves the best performance.

Reliable facial recognition systems are of crucial importance in various applications from entertainment to security. Thanks to the deep-learning concepts introduced in the field, a significant improvement in the performance of the unimodal facial recognition systems has been observed in the recent years. At the same time a multimodal facial recognition is a promising approach. This paper combines the latest successes in both directions by applying deep learning Convolutional Neural Networks (CNN) to the multimodal RGB-D-T based facial recognition problem outperforming previously published results. Furthermore, a late fusion of the CNN-based recognition block with various hand-crafted features (LBP, HOG, HAAR, HOGOM) is introduced, demonstrating even better recognition performance on a benchmark RGB-D-T database. The obtained results in this paper show that the classical engineered features and CNNbased features can complement each other for recognition purposes.