当前位置: X-MOL 学术Comput. Vis. Image Underst. › 论文详情
Our official English website, www.x-mol.net, welcomes your feedback! (Note: you will need to create a separate account there.)
Learning lightweight Multi-Scale Feedback Residual network for single image super-resolution
Computer Vision and Image Understanding ( IF 4.5 ) Pub Date : 2020-06-02 , DOI: 10.1016/j.cviu.2020.103005
Wenjie Xu , Huihui Song , Kaihua Zhang , Qingshan Liu , Jia Liu

In the past years, convolutional neural networks (CNNs) have demonstrated great success for single image super-resolution (SISR). However, existing CNNs for SISR generally have two limitations: (1) the network depth is very deep, which not only weakens the information flow from bottom to top but also has a heavy model capacity; (2) the network architectures are often feed-forward, which prevent the previous layers capturing the useful information from the following layers, limiting the feature learning capability. To address these issues, this paper presents a lightweight Multi-scale Feedback Residual network for SISR. Specifically, we design a lightweight feedback-based recurrent neural network (FRNN) tailored to SISR. The FRNN is consists of a series of recursive Densely-Connected Blocks (DCBs) with the Low-Resolution (LR) image features and the output of the former DCB as inputs. Each DCB adaptively fuses multi-level features from the side-output intermediate feature maps to generate a powerful feature representation. Meanwhile, the DCB cascades a set of Multi-scale Residual Blocks (MRBs), each of which has an enlarged field of view to fully capture multi-scale context information. Moreover, the MRB has a novel Multi-Kernel Fusion Block (MKFB) design, which can dynamically adjust the receptive field size of the output feature representation based on the multi-scale inputs. The whole network of our MFRSR is lightweight with only 4.5M parameters, but achieves favorable performance on five benchmark datasets compared to the state-of-the-art methods in terms of PSNR and SSIM.



中文翻译:

学习轻量级多尺度反馈残差网络以实现单图像超分辨率

在过去的几年中,卷积神经网络(CNN)已经证明了单图像超分辨率(SISR)的巨大成功。但是,现有的用于SISR的CNN通常有两个局限性:(1)网络深度很深,不仅削弱了从下到上的信息流,而且模型容量大。(2)网络架构通常是前馈的,这阻止了前一层从后一层捕获有用信息,从而限制了特征学习能力。为了解决这些问题,本文提出了一种用于SISR的轻量级多尺度反馈残差网络。具体来说,我们设计了针对SISR的轻量级基于反馈的递归神经网络(FRNN)。FRNN由一系列具有低分辨率(LR)图像特征的递归密集连接块(DCB)和以前的DCB的输出作为输入组成。每个DCB自适应地融合来自侧面输出中间特征图的多级特征,以生成功能强大的特征表示。同时,DCB级联了一组多尺度残差块(MRB),每个块具有更大的视野,可以完全捕获多尺度上下文信息。此外,MRB具有新颖的多核融合块(MKFB)设计,该设计可以基于多尺度输入动态调整输出特征表示的接收场大小。我们的MFRSR的整个网络都是轻量级的,只有 每个DCB自适应地融合来自侧面输出中间特征图的多级特征,以生成功能强大的特征表示。同时,DCB级联了一组多尺度残差块(MRB),每个块具有更大的视野,可以完全捕获多尺度上下文信息。此外,MRB具有新颖的多核融合块(MKFB)设计,该设计可以基于多尺度输入动态调整输出特征表示的接收场大小。我们的MFRSR的整个网络都是轻量级的,只有 每个DCB自适应地融合来自侧面输出中间特征图的多级特征,以生成功能强大的特征表示。同时,DCB级联了一组多尺度残差块(MRB),每个块具有更大的视野,可以完全捕获多尺度上下文信息。此外,MRB具有新颖的多核融合块(MKFB)设计,该设计可以基于多尺度输入动态调整输出特征表示的接收场大小。我们的MFRSR的整个网络都是轻量级的,只有 可以根据多尺度输入动态调整输出要素表示的接受域大小。我们的MFRSR的整个网络都是轻量级的,只有 可以根据多尺度输入动态调整输出要素表示的接受场大小。我们的MFRSR的整个网络都是轻量级的,只有45M个参数,但在五个基准数据集上,与PSNR和SSIM方面的最新方法相比,具有良好的性能。

更新日期:2020-06-02
down
wechat
bug