Improved RGB‐D‐T based face recognition. Issue 4 (2nd June 2016)

• Record Type: Journal Article
• Authors: Simón, Marc Oliu; Corneanu, Ciprian; Nasrollahi, Kamal; Nikisins, Olegs; Escalera, Sergio; Sun, Yunlian; Li, Haiqing; Sun, Zhenan; Moeslund, Thomas B.; Greitans, Modris
• Is Part Of: IET biometrics. Volume 5:Issue 4(2016)
• Page Start: 297
• Page End: 303
• Publication Date: 2016-06-02
• Subjects: face recognition -- learning (artificial intelligence) -- neural nets -- visual databases -- image colour analysis; RGB‐D‐T database -- histogram of Gabor ordinal measures -- Haar‐like rectangular features -- histogram of oriented gradients -- local binary patterns -- handcrafted features -- CNN‐based recognition block -- multimodal RGB‐depth‐thermal based facial recognition -- deep learning convolutional neural networks -- multimodal facial recognition -- unimodal facial recognition systems -- improved RGB‐D‐T based face recognition; Biometric identification -- Periodicals; 570.15195
• DOI: 10.1049/iet-bmt.2015.0057
• Languages: English
• ISSNs: 2047-4938

Abstract

Abstract : 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 study combines the latest successes in both directions by applying deep learning convolutional neural networks (CNN) to the multimodal RGB, depth, and thermal (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 (local binary patterns, histograms of oriented gradients, Haar‐like rectangular features, histograms of Gabor ordinal measures) is introduced, demonstrating even better recognition performance on a benchmark RGB‐D‐T database. The obtained results in this study show that the classical engineered features and CNN‐based features can complement each other for recognition purposes.