A Deep Journey into Super-resolution

Abstract

Deep convolutional networks–based super-resolution is a fast-growing field with numerous practical applications. In this exposition, we extensively compare more than 30 state-of-the-art super-resolution Convolutional Neural Networks (CNNs) over three classical and three recently introduced challenging datasets to benchmark single image super-resolution. We introduce a taxonomy for deep learning–based super-resolution networks that groups existing methods into nine categories including linear, residual, multi-branch, recursive, progressive, attention-based, and adversarial designs. We also provide comparisons between the models in terms of network complexity, memory footprint, model input and output, learning details, the type of network losses, and important architectural differences (e.g., depth, skip-connections, filters). The extensive evaluation performed shows the consistent and rapid growth in the accuracy in the past few years along with a corresponding boost in model complexity and the availability of large-scale datasets. It is also observed that the pioneering methods identified as the benchmarks have been significantly outperformed by the current contenders. Despite the progress in recent years, we identify several shortcomings of existing techniques and provide future research directions towards the solution of these open problems. Datasets and codes for evaluation are publicly available at https://github.com/saeed-anwar/SRsurvey.

Does Deep Super-Resolution Enhance UAV Detection?

“…Super-resolution (SR) technique has been studied for many years to increase the image resolution, while preserving fidelity in terms of quality [3,15]. Recently, much research on the field has been focused on exploiting DNNs that have proved to be superior in performance [19,1,6], when compared to old-fashion SR techniques.…”

Image Super-Resolution Reconstruction Using Generative Adversarial Networks Based on Wide-Channel Activation

“…The model in [39] uses a wide-channel residual block similar to DRRN. ESRGAN ranked first in the PIRM2018-SR Challenge competition and obtained a higher perceived quality index than both EDSR and DRRN [47]. The proposed models, called WDSRGAN16 and WDSR-GAN22, and the other five models were tested on the Set14, BSDS200, and Urban100 datasets (with a magnification of 4).…”

Single-Image Super-Resolution Analysis in DCT Spectral Domain

“…Performance metrics. For quantitative evaluation of the SR results, we use the Peak Signal-to-Noise Ratio (PSNR) and Structural Similarity Index (SSIM) [28] evaluation metrics, following the common practice [7]. We note that while higher PSNR and SSIM values are desirable in theory, these metrics are not fully correlated with true perceptual quality [14].…”