logo Idiap Research Institute        
 [BibTeX] [Marc21]
Perspectives and limitations of visible-thermal image pair synthesis via generative adversarial networks
Type of publication: Conference paper
Citation: Panchard_SPIE_2021
Publication status: Published
Booktitle: Security + Defence, Target and Background Signatures VII, Proc. of SPIE
Volume: 11865
Year: 2021
Month: September
Pages: 1186509-1--1186509-8
Publisher: SPIE
Location: online only
URL: https://doi.org/10.1117/12.259...
DOI: 10.1117/12.2599889
Abstract: Many applications rely on thermal imagers to complement or replace visible light sensors in difficult imaging conditions. Recent advances in machine learning have opened the possibility of analyzing or enhancing images, yet these methods require large annotated databases. Training approaches that leverage data augmentation via simulated and synthetically-generated images could offer promising prospects. Here, we report on a method that uses generative adversarial nets (GANs) to synthesize images of a complementary contrast. Starting from a dual-modality dataset of co-registered visible and thermal images, we trained a GAN to generate synthetic thermal images from visible images and vice versa. Our results show that the procedure yields sharp synthesized images that might be used to augment dual-modality datasets or assist in visual interpretation, yet are also subject to the limitations imposed by contrast independence between thermal and visible images.
Keywords: Generative Adversarial Networks, image synthesis, thermal imaging
Projects Idiap
SRML
SRML 21
Authors Panchard, Danick
Marelli, Fran├žois
De Moura Presa, Edouard
Wellig, Peter
Liebling, Michael
Added by: [UNK]
Total mark: 0
Attachments
  • Panchard_SPIE_2021.pdf
Notes