Fast and Accurate Image Super Resolution by Deep CNN with Skip Connection and Network in Network

Abstract

We propose a highly efficient and faster Single Image Super-Resolution (SISR) model with Deep Convolutional neural networks (Deep CNN). Deep CNN have recently shown that they have a significant reconstruction performance on single-image super-resolution. The current trend is using deeper CNN layers to improve performance. However, deep models demand larger computation resources and are not suitable for network edge devices like mobile, tablet and IoT devices. Our model achieves state-of-the-art reconstruction performance with at least 10 times lower calculation cost by Deep CNN with Residual Net, Skip Connection and Network in Network (DCSCN). A combination of Deep CNNs and Skip connection layers are used as a feature extractor for image features on both local and global areas. Parallelized 1x1 CNNs, like the one called Network in Network, are also used for image reconstruction.That structure reduces the dimensions of the previous layer's output for faster computation with less information loss, and make it possible to process original images directly. Also we optimize the number of layers and filters of each CNN to significantly reduce the calculation cost. Thus, the proposed algorithm not only achieves stateof-the-art performance but also achieves faster and more efficient computation. Code is available at https://github.com/jiny2001/dcscn-super-resolution.

DFuseNet: Deep Fusion of RGB and Sparse Depth Information for Image Guided Dense Depth Completion

“…A comprehensive review of these methods is presented by Yang et al [22]. We note that multi-scale architectures with multiple skip connections have been successfully used for image and depth upsampling tasks [23], [24]. Content-aware completion is motivated by a similar problem of learning complete representations from incomplete input data.…”

23‐3: Deep Tone‐mapped HDRNET for High Dynamic Range Image Restoration

“…We use a convolution neural network (CNN) based network to generate tone-mapped HDR images from a given single LDR image. The proposed network is based on the DCSCN [11] architecture. DCSCN is one of the neural network models used in another image restoration area, super resolution, and is known to produce good results despite the small amount of computation.…”

A Hybrid GA-PSO Method for Evolving Architecture and Short Connections of Deep Convolutional Neural Networks

“…Each layer can be connected to all of the forward layers of the same block, which is called densely-connected structure. Such shortcut connections have been heavily investigated in recent years with different variants [10] [13]. However, such shortcut connections are manually designed and there still are a large number of open questions.…”