TVA_GAN (3).pdf (11.84 MB)
TVA-GAN: Attention Guided Generative Adversarial Network For Thermal To Visible Image Transformations
preprint
posted on 2022-01-28, 21:08 authored by NAND YADAVNAND YADAV, Satish Kumar Singh, Shiv Ram DubeyIn the recent advancement of machine learning
methods for realistic image generation and image translation,
Generative Adversarial Networks (GANs) play a vital role. GAN
generates novel samples that look indistinguishable from the real
images. The image translation using a generative adversarial
network refers to unsupervised learning. In this paper, we
translate the thermal images into visible images. Thermal to
Visible image translation is challenging due to the non-availability
of accurate semantic information and smooth textures. The
thermal images contain only single-channel, holding only the
images’ luminance with less feature. We develop a new Cyclic
Attention-based Generative Adversarial Network for Thermal
to Visible Face transformation (TVA-GAN) by incorporating a
new attention-based network. We use attention guidance with
a recurrent block through an Inception module to reduce the
learning space towards the optimum solution.
History
Email Address of Submitting Author
pis2016004@iiita.ac.inSubmitting Author's Institution
Indian Institute of Information Technology AllahabadSubmitting Author's Country
- India