EvoIQA - Explaining Image Distortions with Evolved White-Box Logic

TL;DR

EvoIQA introduces an explainable, genetic programming-based framework that evolves human-readable formulas for image quality assessment, matching deep learning performance while maintaining interpretability.

Contribution

It presents a novel symbolic regression approach for IQA that produces interpretable models, bridging the gap between traditional and deep learning methods.

Findings

Evolved models outperform traditional hand-crafted IQA metrics.

Models achieve performance parity with state-of-the-art deep learning IQA models.

The framework produces human-readable mathematical formulas for image quality assessment.

Abstract

Traditional Image Quality Assessment (IQA) metrics typically fall into one of two extremes: rigid, hand-crafted mathematical models or "black-box" deep learning architectures that completely lack interpretability. To bridge this gap, we propose EvoIQA, a fully explainable symbolic regression framework based on Genetic Programming that Evolves explicit, human-readable mathematical formulas for image quality assessment (IQA). Utilizing a rich terminal set from the VSI, VIF, FSIM, and HaarPSI metrics, our framework inherently maps structural, chromatic, and information-theoretic degradations into observable mathematical equations. Our results demonstrate that the evolved GP models consistently achieve strong alignment between the predictions and human visual preferences. Furthermore, they not only outperform traditional hand-crafted metrics but also achieve performance parity with complex, state-of-the-art deep learning models like DB-CNN, proving that we no longer have to sacrifice interpretability for state-of-the-art performance.