Action recognition using saliency learned from recorded human gaze

作者：

Highlights：

• 3D CNN action features learned using gaze information outperform handcrafted features.

• Using latent variables that localize the action in an SVM framework is beneficial.

• Using saliency learned from gaze in a latent SVM framework is beneficial.

摘要

•3D CNN action features learned using gaze information outperform handcrafted features.•Using latent variables that localize the action in an SVM framework is beneficial.•Using saliency learned from gaze in a latent SVM framework is beneficial.

论文关键词：Action recognition,Saliency,Support Vector Machine (SVM),Latent variable,3D Convolutional Neural Network (3D CNN)

论文评审过程：Received 18 May 2015, Revised 17 December 2015, Accepted 15 June 2016, Available online 27 June 2016, Version of Record 13 July 2016.

论文官网地址：https://doi.org/10.1016/j.imavis.2016.06.006