arXivDaily arXiv每日学术速递 周一至周五更新
arXiv周末暂无论文更新,休息一下吧,周末愉快~~
arXiv 1904.06252cs.CV
收藏

MAANet:图像超分辨率的多视角感知注意网络

MAANet: Multi-view Aware Attention Networks for Image Super-Resolution

Hong Kong Polytechnic University(香港理工大学);Shanghai Jiao Tong University(上海交通大学)

Jingcai Guo, Shiheng Ma, Song Guo

摘要:

近年来,基于深度卷积神经网络(DCNN)的图像超分辨率(SR)在多媒体和计算机视觉领域获得了越来越多的关注,其重点是从低分辨率(LR)图像恢复高分辨率(HR)图像。然而,基于DCNN的方法的一个不可忽视的缺陷是,它们中的大多数不能从具有低频信息冗余的低分辨率图像恢复包含足够高频信息的高分辨率图像。更糟糕的是,随着DCNN深度的增加,训练很容易遇到梯度消失的问题,这使得训练变得更加困难。这些问题阻碍了DCNN在图像SR任务中的有效性。为了解决这些问题,我们提出了多视图感知注意力网络(MAANet)的图像SR任务。具体来说,我们提出了本地意识(LA)和全球意识(GA)的注意力处理LR功能在不平等的方式,它可以突出的高频成分和区分每个功能从LR图像在本地和全球的意见,分别。此外,我们提出了本地注意剩余密集(LARD)块,它结合了LA注意与多个剩余和密集的连接,以适应更深但易于训练的架构。实验结果表明,我们提出的方法可以实现显着的性能相比,其他国家的最先进的方法。

英文摘要:

In most recent years, deep convolutional neural networks (DCNNs) based image super-resolution (SR) has gained increasing attention in multimedia and computer vision communities, focusing on restoring the high-resolution (HR) image from a low-resolution (LR) image. However, one nonnegligible flaw of DCNNs based methods is that most of them are not able to restore high-resolution images containing sufficient high-frequency information from low-resolution images with low-frequency information redundancy. Worse still, as the depth of DCNNs increases, the training easily encounters the problem of vanishing gradients, which makes the training more difficult. These problems hinder the effectiveness of DCNNs in image SR task. To solve these problems, we propose the Multi-view Aware Attention Networks (MAANet) for image SR task. Specifically, we propose the local aware (LA) and global aware (GA) attention to deal with LR features in unequal manners, which can highlight the high-frequency components and discriminate each feature from LR images in the local and the global views, respectively. Furthermore, we propose the local attentive residual-dense (LARD) block, which combines the LA attention with multiple residual and dense connections, to fit a deeper yet easy to train architecture. The experimental results show that our proposed approach can achieve remarkable performance compared with other state-of-the-art methods.

机构、总结、中文摘要均由AI分析而来,可能存在错误,请以论文正文为准。

↑