Explaining Image Enhancement Black-Box Methods through a Path Planning Based Algorithm

TL;DR

This paper introduces eXIE, a path planning algorithm that explains image enhancement black-box methods step-by-step, improving interpretability without sacrificing performance.

Contribution

The paper presents eXIE, a novel path planning algorithm using A* to interpret and replicate image enhancement processes of black-box models.

Findings

eXIE successfully explains enhancement processes of state-of-the-art models.

Sequences of operators generated by eXIE produce similar enhancement results.

eXIE improves interpretability of image enhancement methods.

Abstract

Nowadays, image-to-image translation methods, are the state of the art for the enhancement of natural images. Even if they usually show high performance in terms of accuracy, they often suffer from several limitations such as the generation of artifacts and the scalability to high resolutions. Moreover, their main drawback is the completely black-box approach that does not allow to provide the final user with any insight about the enhancement processes applied. In this paper we present a path planning algorithm which provides a step-by-step explanation of the output produced by state of the art enhancement methods, overcoming black-box limitation. This algorithm, called eXIE, uses a variant of the A* algorithm to emulate the enhancement process of another method through the application of an equivalent sequence of enhancing operators. We applied eXIE to explain the output of several state-of-the-art models trained on the Five-K dataset, obtaining sequences of enhancing operators able to produce very similar results in terms of performance and overcoming the huge limitation of poor interpretability of the best performing algorithms.