Age and gender recognition in the wild with deep attention

作者：

Highlights：

• A novel feedforward attention mechanism for CNNs is proposed.

• The mechanism increases CNNs robustness to image deformations and clutter.

• The proposed mechanism increases CNNs performance for age and gender recognition.

摘要

•A novel feedforward attention mechanism for CNNs is proposed.•The mechanism increases CNNs robustness to image deformations and clutter.•The proposed mechanism increases CNNs performance for age and gender recognition.

论文关键词：Age recognition,Gender recognition,Deep neural networks,Attention mechanisms

论文评审过程：Received 2 February 2017, Revised 17 May 2017, Accepted 25 June 2017, Available online 1 July 2017, Version of Record 17 August 2017.

论文官网地址：https://doi.org/10.1016/j.patcog.2017.06.028