Head pose estimation in the wild using Convolutional Neural Networks and adaptive gradient methods

作者：

Highlights：

• A convolutional neural network approach for head pose estimation is proposed.

• The performance of different network architectures has been measured.

• The use of adaptive gradient methods leads to the state-of-the-art in wild datasets.

• We release a library based on our work which is available under open source licence.

摘要

•A convolutional neural network approach for head pose estimation is proposed.•The performance of different network architectures has been measured.•The use of adaptive gradient methods leads to the state-of-the-art in wild datasets.•We release a library based on our work which is available under open source licence.

论文关键词：Convolutional Neural Networks,Head pose estimation,Adaptive gradient,Deep learning

论文评审过程：Received 11 September 2016, Revised 28 March 2017, Accepted 1 June 2017, Available online 3 June 2017, Version of Record 10 June 2017.

论文官网地址：https://doi.org/10.1016/j.patcog.2017.06.009