A Novel Deep Network Architecture for Reconstructing RGB Facial Images from Thermal for Face Recognition

Andre Litvin, Kamal Nasrollahi, Cagri Ozcinar, Sergio Escalera Guerrero, Thomas B. Moeslund, Gholamreza Anbarjafari

Research output: Contribution to journalJournal articleResearchpeer-review

4 Citations (Scopus)

Abstract

This work proposes a fully convolutional network architecture for RGB face image generation from a given input thermal face image to be applied in face recognition scenarios. The proposed method is based on the FusionNet architecture and increases robustness against overfitting using dropout after bridge connections, randomised leaky ReLUs (RReLUs), and orthogonal regularization. Furthermore, we propose to use a decoding block with resize convolution instead of transposed convolution to improve final RGB face image generation. To validate our proposed network architecture, we train a face classifier and compare its face recognition rate on the reconstructed RGB images from the proposed architecture, to those when reconstructing images with the original FusionNet, as well as when using the original RGB images. As a result, we are introducing a new architecture which leads to a more accurate network.
Original languageEnglish
JournalMultimedia Tools and Applications
Volume78
Issue number18
Pages (from-to)25259-25271
Number of pages13
ISSN1380-7501
DOIs
Publication statusPublished - 30 Sep 2019

Keywords

  • Face recognition
  • Fully convolutional networks
  • FusionNet
  • Thermal imaging

Fingerprint Dive into the research topics of 'A Novel Deep Network Architecture for Reconstructing RGB Facial Images from Thermal for Face Recognition'. Together they form a unique fingerprint.

  • Cite this