Nonlinear Multi-scale Super-resolution Using Deep Learning

Kenneth Tran; Ashkan Panahi; Aniruddha Adiga; Wesam A. Sakla; Hamid Krim

ICASSP 2019accepted0 citations

Nonlinear Multi-scale Super-resolution Using Deep Learning

Kenneth Tran, Ashkan Panahi, Aniruddha Adiga, Wesam A. Sakla, Hamid Krim

Abstract

We propose a deep learning architecture capable of performing up to 8× single image super-resolution. Our architecture incorporates an adversarial component from the super-resolution generative adversarial networks (SRGANs) and a multi-scale learning component from the multiple scale super-resolution network (MSSRNet), which only together can recover smaller structures inherent in satellite images. To further enhance our performance, we integrate progressive growing and training to our network. This, aided by feed forwarding connections in the network to move along and enrich information from previous inputs, produces super-resolved images at scaling factors of 2, 4, and 8. To ensure and enhance the stability of GANs, we employ Wasserstein GANs (WGANs) during training. Experimentally, we find that our architecture can recover small objects in satellite images during super-resolution whereas previous methods cannot.

BibTeX

@inproceedings{icassp2019_nonlinearmultisc,
  title = {Nonlinear Multi-scale Super-resolution Using Deep Learning},
  author = {Kenneth Tran and Ashkan Panahi and Aniruddha Adiga and Wesam A. Sakla and Hamid Krim},
  booktitle = {ICASSP 2019},
  year = {2019}
}