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%.
@article{arxiv.2210.08001,
title = {Learnable Polyphase Sampling for Shift Invariant and Equivariant Convolutional Networks},
author = {Renan A. Rojas-Gomez and Teck-Yian Lim and Alexander G. Schwing and Minh N. Do and Raymond A. Yeh},
journal= {arXiv preprint arXiv:2210.08001},
year = {2022}
}
Comments
Accepted at the Thirty-sixth Conference on Neural Information Processing Systems (NeurIPS 2022)