ML Anthology

Dual Skipping Networks

Changmao Cheng, Yanwei Fu, Yu-Gang Jiang, Wei Liu, Wenlian Lu, Jianfeng Feng, Xiangyang Xue
CVPR 2018
doi:10.1109/CVPR.2018.00428 /cvpr/2018/cheng2018cvpr-dual/

Abstract

Inspired by the recent neuroscience studies on the left-right asymmetry of the human brain in processing low and high spatial frequency information, this paper introduces a dual skipping network which carries out coarse-to-fine object categorization. Such a network has two branches to simultaneously deal with both coarse and fine-grained classification tasks. Specifically, we propose a layer-skipping mechanism that learns a gating network to predict which layers to skip in the testing stage. This layer-skipping mechanism endows the network with good flexibility and capability in practice. Evaluations are conducted on several widely used coarse-to-fine object categorization benchmarks, and promising results are achieved by our proposed network model.

PDF CVPR Semantic Scholar

Cite

Text

Cheng et al. "Dual Skipping Networks." Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, 2018. doi:10.1109/CVPR.2018.00428

Markdown

[Cheng et al. "Dual Skipping Networks." Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, 2018.](https://mlanthology.org/cvpr/2018/cheng2018cvpr-dual/) doi:10.1109/CVPR.2018.00428

BibTeX

@inproceedings{cheng2018cvpr-dual,
  title     = {{Dual Skipping Networks}},
  author    = {Cheng, Changmao and Fu, Yanwei and Jiang, Yu-Gang and Liu, Wei and Lu, Wenlian and Feng, Jianfeng and Xue, Xiangyang},
  booktitle = {Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition},
  year      = {2018},
  doi       = {10.1109/CVPR.2018.00428},
  url       = {https://mlanthology.org/cvpr/2018/cheng2018cvpr-dual/}
}