Adaptive Local Implicit Image Function for Arbitrary-scale Super-resolution

TL;DR

This paper introduces A-LIIF, an adaptive local implicit image function that improves arbitrary-scale super-resolution by capturing local image properties, reducing artifacts, and enhancing high-frequency detail reconstruction.

Contribution

The paper proposes A-LIIF, a novel adaptive local implicit image function that models local image properties for better super-resolution at arbitrary scales.

Findings

A-LIIF outperforms existing methods on multiple benchmarks.

It effectively reconstructs high-frequency textures and structures.

The method reduces artifacts and ringing around edges.

Abstract

Image representation is critical for many visual tasks. Instead of representing images discretely with 2D arrays of pixels, a recent study, namely local implicit image function (LIIF), denotes images as a continuous function where pixel values are expansion by using the corresponding coordinates as inputs. Due to its continuous nature, LIIF can be adopted for arbitrary-scale image super-resolution tasks, resulting in a single effective and efficient model for various up-scaling factors. However, LIIF often suffers from structural distortions and ringing artifacts around edges, mostly because all pixels share the same model, thus ignoring the local properties of the image. In this paper, we propose a novel adaptive local image function (A-LIIF) to alleviate this problem. Specifically, our A-LIIF consists of two main components: an encoder and a expansion network. The former captures cross-scale image features, while the latter models the continuous up-scaling function by a weighted combination of multiple local implicit image functions. Accordingly, our A-LIIF can reconstruct the high-frequency textures and structures more accurately. Experiments on multiple benchmark datasets verify the effectiveness of our method. Our codes are available at \url{https://github.com/LeeHW-THU/A-LIIF}.