Dynamic Implicit Image Function for Efficient Arbitrary-Scale Image Representation

TL;DR

The paper introduces DIIF, a novel implicit image representation method that significantly improves computational efficiency for arbitrary-scale super-resolution, enabling real-time applications while maintaining state-of-the-art quality.

Contribution

DIIF employs coordinate grouping and slicing with a Coarse-to-Fine MLP to reduce computational cost in implicit image representations for arbitrary scales.

Findings

DIIF achieves state-of-the-art super-resolution performance.

DIIF significantly reduces computational cost for arbitrary-scale SR.

DIIF enables real-time arbitrary-scale image representation.

Abstract

Recent years have witnessed the remarkable success of implicit neural representation methods. The recent work Local Implicit Image Function (LIIF) has achieved satisfactory performance for continuous image representation, where pixel values are inferred from a neural network in a continuous spatial domain. However, the computational cost of such implicit arbitrary-scale super-resolution (SR) methods increases rapidly as the scale factor increases, which makes arbitrary-scale SR time-consuming. In this paper, we propose Dynamic Implicit Image Function (DIIF), which is a fast and efficient method to represent images with arbitrary resolution. Instead of taking an image coordinate and the nearest 2D deep features as inputs to predict its pixel value, we propose a coordinate grouping and slicing strategy, which enables the neural network to perform decoding from coordinate slices to pixel value slices. We further propose a Coarse-to-Fine Multilayer Perceptron (C2F-MLP) to perform decoding with dynamic coordinate slicing, where the number of coordinates in each slice varies as the scale factor varies. With dynamic coordinate slicing, DIIF significantly reduces the computational cost when encountering arbitrary-scale SR. Experimental results demonstrate that DIIF can be integrated with implicit arbitrary-scale SR methods and achieves SOTA SR performance with significantly superior computational efficiency, thereby opening a path for real-time arbitrary-scale image representation. Our code can be found at https://github.com/HeZongyao/DIIF.