SRGAN-CKAN: Expressive Super-Resolution with Nonlinear Functional Operators under Minimal Resources

2026-05-02

AI Summary

This work addresses the severe degradation of high-frequency details in large-scale super-resolution tasks and the prohibitive computational cost of existing high-performance methods, which hinders deployment in resource-constrained scenarios. To this end, we propose SRGAN-CKAN, a novel framework that integrates Convolutional Kolmogorov–Arnold Networks (CKAN) into an adversarial learning paradigm. By replacing conventional linear convolutions with spline-based nonlinear functions, our approach redefines the local image reconstruction operator. This design significantly enhances the model’s capacity to capture high-frequency textures and complex structures at extremely low computational cost, achieving a favorable trade-off between perceptual quality and reconstruction fidelity. Consequently, SRGAN-CKAN enables efficient yet highly expressive super-resolution reconstruction.

Abstract

Single-Image Super-Resolution (SISR) aims to reconstruct a High-Resolution (HR) image from a Low-Resolution (LR) observation, a fundamentally ill-posed problem where high-frequency details are severely degraded at large upscaling factors. Recent advances have been driven by transformer-based architectures and diffusion models improve global context modeling and perceptual quality at the cost of increased computational complexity. In contrast, this work focuses on enhancing the expressivity of local operators under minimal resources. We propose SRGAN--CKAN, a hybrid super-resolution framework that integrates Convolutional Kolmogorov--Arnold Networks (CKAN) into an adversarial learning setting reformulating convolution as a nonlinear patch-based transformation. The proposed operator replaces linear local mappings with spline-based functional representations, allowing expressive modeling of complex local structures and high-frequency textures using minimal hardware resources. Experimental results demonstrate that the proposed approach improves perceptual quality while preserving reconstruction fidelity, achieving a favorable balance between distortion-based and perceptual metrics. These results are obtained under constrained computational settings, highlighting the efficiency of the proposed formulation. Overall, this work introduces a complementary direction to existing approaches by improving the representational power of local transformations, providing an efficient and scalable alternative to globally intensive architectures.

Problem

• Single-Image Super-Resolution
• High-Frequency Details
• Minimal Resources
• Ill-Posed Problem
• Perceptual Quality

Innovation

• Convolutional Kolmogorov–Arnold Networks
• Nonlinear Functional Operators
• Single-Image Super-Resolution
• Perceptual Quality
• Resource-Efficient Learning