Deep Residual Dense Network for Single Image Super-Resolution

Abstract

In this paper, we propose a deep residual dense network (DRDN) for single image super- resolution. Based on human perceptual characteristics, the residual in residual dense block strategy (RRDB) is exploited to implement various depths in network architectures. The proposed model exhibits a simple sequential structure comprising residual and dense blocks with skip connections. It improves the stability and computational complexity of the network, as well as the perceptual quality. We adopt a perceptual metric to learn and assess the quality of the reconstructed images. The proposed model is trained with the Diverse2k dataset, and the performance is evaluated using standard datasets. The experimental results confirm that the proposed model exhibits superior performance, with better reconstruction results and perceptual quality than conventional methods.

Super‐resolution using deep residual network with spectral normalization

“…For model evaluation, the training and testing datasets were used separately. Finally, perceptual quality was evaluated using two metrics: the universal image quality index (UQI) [21, 22], a reference‐image‐based metric, and the perception‐based image quality evaluator (PIQE) [23], a non‐reference‐image metric.…”

“…Set 5 [23] Set 14 [23] BSD 100 [ proposed model performed the best on all datasets for all scale factors. A visual comparison of the models is shown in Figures 3 and 4 for set 5, set 14, BSD100, and Urban 100, respectively.…”

“…The specifications of the proposed model are listed in Table 1. The proposed model reduced the parameters by 20 times compared to EDSR [11], six times compared to DRDN [23], and three times compared to MDSR [11].…”

A residual dense comprehensively regulated convolutional neural network to identify spectral information for egg quality traceability

“…Kim et al 39 proposed a grouped residual dense network (RDN) for image denoising. Musunuri et al 40 proposed a deep RDN for single image super-resolution. These methods took full advantage of the hierarchical features from all the convolution layers, obtaining high classification accuracy and stability.…”

Multi-Path Deep CNN with Residual Inception Network for Single Image Super-Resolution

“…They adopted the generator part of the super-resolution generative adversarial network (SRGAN) as the model structure and employed the residual connections between layers. Musunuri et al [66] introduced the concept of deep residual dense network architecture for single image super-resolution abbreviated as DRDN. The network architecture is based on the combination of residual and dense blocks with skip connections.…”