Color-UNet++: A resolution for colorization of grayscale images using improved UNet++

Color-UNet++: A resolution for colorization of grayscale images using improved UNet++
复制标题

Color-UNet:使用改进的 UNet 进行灰度图像着色的分辨率

DOI:
10.1007/s11042-021-10830-2
复制
发表时间:
2021-03
影响因子:
3.6
通讯作者:
Qing Duan
Qing Duan
中科院分区:
计算机科学4区
文献类型:
--
作者:
Yide Di;Xiaoke Zhu;Xin Jin;Qiwei Dou;Wei Zhou;Qing Duan

文献摘要

参考文献

相似文献

着色是计算机辅助的颜色应用于灰度图像,这给现代基于深度学习的方法带来了两个问题。一个是提供具有高表达性和强学习能力的着色模型,因为现有模型难以同时擅长着色和易于训练。另一种是返回没有不均匀重叠的图片。本文提出了一个名为Color-UNet++的深度卷积网络框架,用于解决这些着色问题。Color-UNet++通过在反向传播过程中捕获更多的传输和中间结果来调整梯度分散和爆炸。我们调整了去卷积结构,以解决不均匀重叠的问题。我们在YUV而不是RGB颜色空间中设计模型,目标函数适合着色问题,可以捕获广泛的颜色。在LFW和LSUN数据集上的大量实验结果证实了该方法的优越性。
Colorization is the computer-assisted application of color to a gray scale image, which presents two problems to modern deep learning-based approaches. One is to provide colorization models with both high expressibility and strong learning ability, as current models have difficulty both excelling at coloring and being easy to train. The other is to return a picture without uneven overlap. This paper proposes a deep convolutional network framework called Color-UNet++ for the end-to-end solution of these colorization problems. Color-UNet++ is adjusted to settle gradient dispersion and explosion by capturing more transfer and intermediate results during backpropagation. We adjust the de-convolution structure to solve the problem of uneven overlap. We design the model in YUV instead of RGB color space, with an objective function that is appropriate to the coloring problem and can capture a wide range of colors. A large number of experimental results on LFW and LSUN datasets confirm the method’s superiority.
DOI: --
发表时间: 2012-12
期刊: --
影响因子: --
作者:
Gary B. Huang;Marwan A. Mattar;Honglak Lee;E. Learned-Miller
通讯作者: Gary B. Huang;Marwan A. Mattar;Honglak Lee;E. Learned-Miller
基于精确四元数极调和傅立叶矩的几何不变彩色医学图像无水印
DOI: 10.1109/access.2019.2935174
发表时间: 2019-08
期刊: IEEE Access
影响因子: 3.9
作者:
Xia Zhiqiu;Wang Xingyuan;Wang Mingxu;Unar Salahuddin;Wang Chunpeng;Liu Ying;Li Xiaoxiao
通讯作者: Li Xiaoxiao
DOI: 10.1016/j.entcom.2019.100298
发表时间: 2019-05
期刊: Entertain. Comput.
影响因子: --
作者:
Samad Roohi;Aynaz Forouzandeh
通讯作者: Samad Roohi;Aynaz Forouzandeh
DOI: 10.5220/0005676401260133
发表时间: 2016
期刊: --
影响因子: --
作者:
Van Nguyen;Vicky Sintunata;T. Aoki
通讯作者: Van Nguyen;Vicky Sintunata;T. Aoki
DOI: 10.1007/978-3-319-64689-3_15
发表时间: 2017-08
期刊: --
影响因子: --
作者:
D. Varga;T. Szirányi
通讯作者: D. Varga;T. Szirányi