Extreme Network Compression via Filter Group Approximation

Bo Peng, Wenming Tan, Zheyang Li, Shun Zhang, Di Xie, Shiliang Pu

ECCV 2018

doi:10.1007/978-3-030-01237-3_19 /eccv/2018/peng2018eccv-extreme/

Abstract

In this paper we propose a novel decomposition method based on filter group approximation, which can significantly reduce the redundancy of deep convolutional neural networks (CNNs) while maintaining the majority of feature representation. Unlike other low-rank decomposition algorithms which operate on spatial or channel dimension of filters, our proposed method mainly focuses on exploiting the filter group structure for each layer. For several commonly used CNN models, including VGG and ResNet, our method can reduce over 80% floating-point operations (FLOPs) with less accuracy drop than state-of-the-art methods on various image classification datasets. Besides, experiments demonstrate that our method is conducive to alleviating degeneracy of the compressed network, which hurts the convergence and performance of the network.

PDF ECCV Semantic Scholar

Cite

Text

Peng et al. "Extreme Network Compression via Filter Group Approximation." Proceedings of the European Conference on Computer Vision (ECCV), 2018. doi:10.1007/978-3-030-01237-3_19

Markdown

[Peng et al. "Extreme Network Compression via Filter Group Approximation." Proceedings of the European Conference on Computer Vision (ECCV), 2018.](https://mlanthology.org/eccv/2018/peng2018eccv-extreme/) doi:10.1007/978-3-030-01237-3_19

BibTeX

@inproceedings{peng2018eccv-extreme,
  title     = {{Extreme Network Compression via Filter Group Approximation}},
  author    = {Peng, Bo and Tan, Wenming and Li, Zheyang and Zhang, Shun and Xie, Di and Pu, Shiliang},
  booktitle = {Proceedings of the European Conference on Computer Vision (ECCV)},
  year      = {2018},
  doi       = {10.1007/978-3-030-01237-3_19},
  url       = {https://mlanthology.org/eccv/2018/peng2018eccv-extreme/}
}