Learnable Polyphase Sampling for Shift Invariant and Equivariant Convolutional Networks

Renan A. Rojas-Gomez · Teck-Yian Lim · Alex Schwing · Minh Do · Raymond A. Yeh

Hall J #912

Keywords: [ convolutional neural networks ] [ Polyphase Decomposition ] [ Shift equivariance ] [ Shift invariance ]

[ Abstract ]
[ Paper [ Poster [ OpenReview
Tue 29 Nov 2 p.m. PST — 4 p.m. PST


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%.

Chat is not available.