Efficient adaptive inference for deep convolutional neural networks using hierarchical early exits

作者：

Highlights：

• An efficient hierarchical aggregation scheme for early exits is proposed.

• A classification layer reuse approach is used to further reduce the complexity.

• An adaptive classification approach is proposed further improving the performance.

• The effectiveness of the proposed method is demonstrated using four image datasets.

摘要

•An efficient hierarchical aggregation scheme for early exits is proposed.•A classification layer reuse approach is used to further reduce the complexity.•An adaptive classification approach is proposed further improving the performance.•The effectiveness of the proposed method is demonstrated using four image datasets.

论文关键词：Adaptive inference,Early exits,Bag-of-Features,Deep convolutional neural networks,Hierarchical representations

论文评审过程：Received 13 November 2019, Revised 9 March 2020, Accepted 25 March 2020, Available online 11 May 2020, Version of Record 11 May 2020.

论文官网地址：https://doi.org/10.1016/j.patcog.2020.107346