Normalization methods are essential components in convolutional neural networks (CNNs). They either standardize or whiten data using statistics estimated in predefined sets of pixels. Unlike existing works that design normalization techniques for specific tasks, we propose Switchable Whitening (SW), which provides a general form unifying different whitening methods as well as standardization methods. SW learns to switch among these operations in an end-to-end manner. It has several advantages. First, SW adaptively selects appropriate whitening or standardization statistics for different tasks (see Fig.1), making it well suited for a wide range of tasks without manual design. Second, by integrating benefits of different normalizers, SW shows consistent improvements over its counterparts in various challenging benchmarks. Third, SW serves as a useful tool for understanding the characteristics of whitening and standardization techniques. We show that SW outperforms other alternatives on image classification (CIFAR-10/100, ImageNet), semantic segmentation (ADE20K, Cityscapes), domain adaptation (GTA5, Cityscapes), and image style transfer (COCO). For example, without bells and whistles, we achieve state-of-the-art performance with 45.33% mIoU on the ADE20K dataset. Code is available at https://github.com/XingangPan/Switchable-Whitening.
翻译:正常化方法是进化神经网络(CNNs)的基本组成部分。 它们要么标准化, 要么白化数据, 使用预先定义的像素组中估计的统计数据。 与设计具体任务正常化技术的现有工作不同, 我们提议可转换白化( SW), 它提供了一种统一不同白化方法以及标准化方法的一般形式。 学会以端到端的方式在这些操作中转换。 它有几个优点。 首先, SW适应性地选择了不同任务的适当白化或标准化统计数据(见Fig.1), 使这些数据完全适合各种任务,而没有手工设计。 其次, 通过整合不同正常化者的利益, SW在各种具有挑战性的基准中显示相对于对应者的一致的改进。 第三, SW是了解白化和标准化技术特点的有用工具。 我们显示, SW在图像分类( CIFAR- 10- 100, 图像网)、 语系分化( ADE20K, 城市景区)、 域适应(GTA5, 城市景景) 和图像样式转移(CO) 。 例如, 没有Bells and 和口哨, 我们在M33/ KASet- ASet- AS. we a. we is ad. sal- a.