ID 原文 译文
40046 最后，对基于深度学习的三维模型检索算法目前存在的问题进行分析和讨论，并展望未来发展的新方向。 Finally, we analyze and discuss existing issues of deep learning based 3 D model retrieval methods, and predict possible promising directions for this research topic.
40047 随着动漫产业的快速发展，动漫人脸的生成成为一项关键技术。 With the rapid development of the animation industry, the face generation with animation characters becomes a key technology.
40048 由于动漫人脸具有的高度简化和抽象的独特风格以及倾向于具有清晰的边缘、平滑的阴影和相对简单的纹理，现有方法中的损失函数面临很大的挑战，同时绘画的风格迁移技术无法获得满意的动漫结果。 The existing style transfer technology of painting style cannot obtain satisfactory animation results due to the following characteristics of animation:(1)Animation has a highly simplified and abstract unique style, and(2)animation tends to have clear edges and smooth shadows and relatively simple textures, which poses great challenges to the loss function in existing methods.
40049 因此，本文提出了一种新颖的适用于动漫图像的损失函数， This paper proposes a novel loss function suitable for animation.
40050 该函数的语义损失表示为VGG网络高级特征图中的正则化形式，以应对真实图像和动漫图像之间不同的风格，具有的边缘增强的边缘清晰损失可以保留动漫图像的边缘清晰度。 In the loss function, the semantic loss is expressed as a regularized form in the high-level feature map of the VGG network to deal with the different styles between real and animation images, and the edge sharpness loss with edge enhancement can preserve the edge sharpness of animation images.
40051 4个公开数据集上的实验表明：通过本文提出的损失函数可以生成清晰生动的动漫人脸图像； Experiments on the four public data sets show that through the proposed loss function, clear and vivid animation images can be generated.
40052 在CK+数据集中，本文方法相比于现有的方法识别率提高了0.43%（宫崎骏风格）和3.29%（新海诚风格）；在RAF数据集中，本文方法识别率提高了0.85%（宫崎骏风格）和2.42%（新海诚风格）；在SFEW数据集中，本文方法识别率提高了0.71%（宫崎骏风格）和3.14%（新海诚风格）； Moreover, in the CK+ data set, the recognition rate of the proposed method is increased by 0.43%(Miyazaki Hayao style)and 3.29%(Makoto Shinkai style)compared with the existing method, increased by 0.85%(Miyazaki Hayao style)and 2.42%(Makoto Shinkai style)in the RAF data set, and increased by 0.71%(Miyazaki Hayao style)and 3.14%(Makoto Shinkai style) in the SFEW data set, respectively.
40053 在Celeba数据集中也显示了本文方法优异的生成效果。 The generation effect in the Celeba data set is also demonstrated.
40054 实验结果说明本文方法结合了深度学习模型的优点，使检测结果更加准确。 The above results show that the proposed method combines the advantages of the deep learning model to make the detection result more accurate.
40055 由于水下显著性检测数据集不足，导致基于深度学习的水下图像显著性检测网络容易出现过拟合的问题，从而影响显著性检测网络的性能。 Due to the insufficient underwater salient object detection datasets, the underwater image salient object detection network based on deep learning is prone to overfitting, which affects the performance of the network.