FreqINR: Frequency Consistency for Implicit Neural Representation with Adaptive DCT Frequency Loss

TL;DR

FreqINR introduces a frequency consistency approach with adaptive DCT loss to improve arbitrary-scale super-resolution, enhancing detail and spectral coherence while maintaining computational efficiency.

Contribution

It proposes a novel frequency consistency method with adaptive DCT loss for implicit neural representations, improving super-resolution quality at arbitrary scales.

Findings

Achieves state-of-the-art super-resolution performance.

Improves spectral coherence between LR and HR images.

Offers a lightweight and efficient super-resolution method.

Abstract

Recent advancements in local Implicit Neural Representation (INR) demonstrate its exceptional capability in handling images at various resolutions. However, frequency discrepancies between high-resolution (HR) and ground-truth images, especially at larger scales, result in significant artifacts and blurring in HR images. This paper introduces Frequency Consistency for Implicit Neural Representation (FreqINR), an innovative Arbitrary-scale Super-resolution method aimed at enhancing detailed textures by ensuring spectral consistency throughout both training and inference. During training, we employ Adaptive Discrete Cosine Transform Frequency Loss (ADFL) to minimize the frequency gap between HR and ground-truth images, utilizing 2-Dimensional DCT bases and focusing dynamically on challenging frequencies. During inference, we extend the receptive field to preserve spectral coherence between low-resolution (LR) and ground-truth images, which is crucial for the model to generate high-frequency details from LR counterparts. Experimental results show that FreqINR, as a lightweight approach, achieves state-of-the-art performance compared to existing Arbitrary-scale Super-resolution methods and offers notable improvements in computational efficiency. The code for our method will be made publicly available.