POEM: 1-bit Point-wise Operations based on Expectation-Maximization for Efficient Point Cloud Processing
Abstract
Real-time point cloud processing is fundamental for lots of computer vision tasks, while still challenged by the computational problem on resource-limited edge devices. To address this issue, we implement XNOR-Net-based binary neural networks (BNNs) for an efficient point cloud processing, but its performance is severely suffered due to two main drawbacks, Gaussian-distributed weights and non-learnable scale factor. In this paper, we introduce point-wise operations based on Expectation-Maximization (POEM) into BNNs for efficient point cloud processing. The EM algorithm can efficiently constrain weights for a robust bi-modal distribution. We lead a well-designed reconstruction loss to calculate learnable scale factors to enhance the representation capacity of 1-bit fully-connected (Bi-FC) layers. Extensive experiments demonstrate that our POEM surpasses existing the state-of-the-art binary point cloud networks by a significant margin, up to 6.7 %.
Cite
@article{arxiv.2111.13386,
title = {POEM: 1-bit Point-wise Operations based on Expectation-Maximization for Efficient Point Cloud Processing},
author = {Sheng Xu and Yanjing Li and Junhe Zhao and Baochang Zhang and Guodong Guo},
journal= {arXiv preprint arXiv:2111.13386},
year = {2021}
}
Comments
Accepted by BMVC 2021. arXiv admin note: text overlap with arXiv:2010.05501 by other authors