Showing 1–1 of 1 results for author: Shandeelya, A P
-
Enhancing Perceptual Loss with Adversarial Feature Matching for Super-Resolution
Authors:
Akella Ravi Tej,
Shirsendu Sukanta Halder,
Arunav Pratap Shandeelya,
Vinod Pankajakshan
Abstract:
Single image super-resolution (SISR) is an ill-posed problem with an indeterminate number of valid solutions. Solving this problem with neural networks would require access to extensive experience, either presented as a large training set over natural images or a condensed representation from another pre-trained network. Perceptual loss functions, which belong to the latter category, have achieved…
▽ More
Single image super-resolution (SISR) is an ill-posed problem with an indeterminate number of valid solutions. Solving this problem with neural networks would require access to extensive experience, either presented as a large training set over natural images or a condensed representation from another pre-trained network. Perceptual loss functions, which belong to the latter category, have achieved breakthrough success in SISR and several other computer vision tasks. While perceptual loss plays a central role in the generation of photo-realistic images, it also produces undesired pattern artifacts in the super-resolved outputs. In this paper, we show that the root cause of these pattern artifacts can be traced back to a mismatch between the pre-training objective of perceptual loss and the super-resolution objective. To address this issue, we propose to augment the existing perceptual loss formulation with a novel content loss function that uses the latent features of a discriminator network to filter the unwanted artifacts across several levels of adversarial similarity. Further, our modification has a stabilizing effect on non-convex optimization in adversarial training. The proposed approach offers notable gains in perceptual quality based on an extensive human evaluation study and a competent reconstruction fidelity when tested on objective evaluation metrics.
△ Less
Submitted 15 May, 2020;
originally announced May 2020.
