Ablation Study to Derive a Computationally Efficient Deep Learning-Based Super-Resolution Approach

Recent research on single-image super-resolution (SR) shows that deep learning-based methods outperform state-of-the-art techniques but at the cost of increased memory consumption and computational complexity. This results in longer training and inference times and higher GPU memory requirements compared to traditional approaches. Network architecture modifications can impact performance, complexity, and memory needs. This paper explores enhancing SR performance using a simple yet efficient deep-learning SR model, focusing on local and global connections in residual networks, channel attention mechanisms, and up-sampling techniques. Our efficient, lightweight, locally dense residual SR architecture achieves performance comparable to state-of-the-art models, reducing spatial complexity by up to 1/6 and inference time by half compared to the baseline.

I. Introduction and Related Work

Single Image Super-Resolution (SR) enhances the resolution of low-resolution images to generate high-resolution counterparts [2]. This process focuses on restoring high-frequency information lost during image acquisition or compression [3], significantly improving visual quality. SR has diverse applications in computer vision, including enhancing medical images, improving surveillance footage, and analyzing satellite imagery [4],