目的针对现有刺绣模拟算法中针线感不强、针线轨迹方向单一等问题,提出了一种基于多尺度双通道卷积神经网络的刺绣模拟算法。方法1)搭建多尺度双通道网络,选取一幅刺绣艺术作品作为风格图像,将MSCOCO(microsoft common objects in conte...目的针对现有刺绣模拟算法中针线感不强、针线轨迹方向单一等问题,提出了一种基于多尺度双通道卷积神经网络的刺绣模拟算法。方法1)搭建多尺度双通道网络,选取一幅刺绣艺术作品作为风格图像,将MSCOCO(microsoft common objects in context)数据集作为训练集,输入网络得到VGG(visual geometry group)网络损失和拉普拉斯损失;2)将总损失值传回到网络,通过梯度下降法更新网络参数,并且重复更新参数直到指定的训练次数完成网络训练;3)选取一幅目标图像作为刺绣模拟的内容图像,输入训练完成的网络,获得具有刺绣艺术风格的结果图像;4)使用掩模图像将得到的结果图像与绣布图像进行图像融合,即完成目标图像的刺绣模拟。结果本文算法能产生明显的针线感和多方向的针线轨迹,增强了刺绣模拟绘制艺术作品的表现力。结论本文将输入图像经过多尺度双通道卷积神经网络进行前向传播,并使用VGG19、VGG16和拉普拉斯模块作为损失网络进行刺绣模拟。实验结果表明,与现有卷积神经网络风格模拟算法对比,本文提出的网络能够学习到刺绣艺术风格图像的针线特征,得到的图像贴近真实刺绣艺术作品。展开更多
To simulate a multivariate density with multi_hump, Markov chainMonte Carlo method encounters the obstacle of escaping from one hump to another, since it usually takes extraordinately long time and then becomes practi...To simulate a multivariate density with multi_hump, Markov chainMonte Carlo method encounters the obstacle of escaping from one hump to another, since it usually takes extraordinately long time and then becomes practically impossible to perform. To overcome these difficulties, a reversible scheme to generate a Markov chain, in terms of which the simulated density may be successful in rather general cases of practically avoiding being trapped in local humps, was suggested.展开更多
文摘目的针对现有刺绣模拟算法中针线感不强、针线轨迹方向单一等问题,提出了一种基于多尺度双通道卷积神经网络的刺绣模拟算法。方法1)搭建多尺度双通道网络,选取一幅刺绣艺术作品作为风格图像,将MSCOCO(microsoft common objects in context)数据集作为训练集,输入网络得到VGG(visual geometry group)网络损失和拉普拉斯损失;2)将总损失值传回到网络,通过梯度下降法更新网络参数,并且重复更新参数直到指定的训练次数完成网络训练;3)选取一幅目标图像作为刺绣模拟的内容图像,输入训练完成的网络,获得具有刺绣艺术风格的结果图像;4)使用掩模图像将得到的结果图像与绣布图像进行图像融合,即完成目标图像的刺绣模拟。结果本文算法能产生明显的针线感和多方向的针线轨迹,增强了刺绣模拟绘制艺术作品的表现力。结论本文将输入图像经过多尺度双通道卷积神经网络进行前向传播,并使用VGG19、VGG16和拉普拉斯模块作为损失网络进行刺绣模拟。实验结果表明,与现有卷积神经网络风格模拟算法对比,本文提出的网络能够学习到刺绣艺术风格图像的针线特征,得到的图像贴近真实刺绣艺术作品。
文摘To simulate a multivariate density with multi_hump, Markov chainMonte Carlo method encounters the obstacle of escaping from one hump to another, since it usually takes extraordinately long time and then becomes practically impossible to perform. To overcome these difficulties, a reversible scheme to generate a Markov chain, in terms of which the simulated density may be successful in rather general cases of practically avoiding being trapped in local humps, was suggested.
