Amortised MAP Inference for Image Super-resolution
Casper Kaae S{\o}nderby, Jose Caballero, Lucas Theis, Wenzhe Shi,, Ferenc Husz\'ar

TL;DR
This paper introduces a neural network-based amortised MAP inference method for image super-resolution, producing more plausible high-resolution images by integrating image priors and efficient inference techniques.
Contribution
It proposes a novel neural network architecture for amortised MAP inference in super-resolution, combining it with three training methods including GANs, to improve image plausibility.
Findings
GAN-based approach outperforms other methods on real images
New neural architecture ensures high-res output consistency with low-res input
Establishes a connection between GANs and amortised variational inference
Abstract
Image super-resolution (SR) is an underdetermined inverse problem, where a large number of plausible high-resolution images can explain the same downsampled image. Most current single image SR methods use empirical risk minimisation, often with a pixel-wise mean squared error (MSE) loss. However, the outputs from such methods tend to be blurry, over-smoothed and generally appear implausible. A more desirable approach would employ Maximum a Posteriori (MAP) inference, preferring solutions that always have a high probability under the image prior, and thus appear more plausible. Direct MAP estimation for SR is non-trivial, as it requires us to build a model for the image prior from samples. Furthermore, MAP inference is often performed via optimisation-based iterative algorithms which don't compare well with the efficiency of neural-network-based alternatives. Here we introduce new…
Peer Reviews
No public reviews on file for this paper yet. If you reviewed it on a platform where reviews are public (OpenReview, ICLR, NeurIPS, ICML), you can paste yours below so the community can read it here.
Videos
No videos yet. Explain this paper in a talk, walkthrough, or lecture? Add one.
Taxonomy
TopicsAdvanced Image Processing Techniques · Image Processing Techniques and Applications · Advanced Vision and Imaging
MethodsConvolution · Dogecoin Customer Service Number +1-833-534-1729
