论文标题
切成薄片的神经纹理综合损失
A Sliced Wasserstein Loss for Neural Texture Synthesis
论文作者
论文摘要
我们根据从优化对象识别的卷积神经网络的特征激活（例如VGG-19）中提取的统计数据（例如，VGG-19）来解决计算纹理损失的问题。基本的数学问题是特征空间中两个分布之间的距离的度量。革兰氏矩阵损失是此问题的无处不在近似，但要遇到几个缺点。我们的目标是促进切成薄片的瓦斯坦距离作为替代。从理论上讲，它是通过优化或训练生成性神经网络在视觉上质量合成的视觉上优越的结果，并且实现的结果在理论上得到证明，实用，易于实现。
We address the problem of computing a textural loss based on the statistics extracted from the feature activations of a convolutional neural network optimized for object recognition (e.g. VGG-19). The underlying mathematical problem is the measure of the distance between two distributions in feature space. The Gram-matrix loss is the ubiquitous approximation for this problem but it is subject to several shortcomings. Our goal is to promote the Sliced Wasserstein Distance as a replacement for it. It is theoretically proven,practical, simple to implement, and achieves results that are visually superior for texture synthesis by optimization or training generative neural networks.
