We propose learnable polyphase sampling (LPS), a pair of learnable down/upsampling layers that enable truly shift-invariant and equivariant convolutional networks. LPS can be trained end-to-end from data and generalizes existing handcrafted downsampling layers. It is widely applicable as it can be integrated into any convolutional network by replacing down/upsampling layers. We evaluate LPS on image classification and semantic segmentation. Experiments show that LPS is on-par with or outperforms existing methods in both performance and shift consistency. For the first time, we achieve true shift-equivariance on semantic segmentation (PASCAL VOC), i.e., 100% shift consistency, outperforming baselines by an absolute 3.3%.

本文提出了可学习的多相采样（LPS），它是一对可学习的下/上采样层，使得卷积网络能够真正具有平移不变性和等变性，可以替换任何卷积网络的下/上采样层，本文在图像分类和语义分割方面对LPS进行了评估，实验表明，LPS在性能和平移一致性方面与现有方法相当或优于现有方法。本文首次在语义分割（PASCAL VOC）中实现了真正的平移等变性，即100％的平移一致性，优于基线3.3％。

可学习多相采样的平移不变和等变卷积网络