Ordered trajectories for human action recognition with large number of classes

作者：

Highlights：

• A technique that captures information of objects with longer duration.

• A feature selection like approach that delivers better performance than several trajectory variants.

• Removal of a large number of trajectories related to background noise.

• We apply our technique on action datasets HMDB51, UCF50 and UCF101 containing largest number of classes till date.

摘要

•A technique that captures information of objects with longer duration.•A feature selection like approach that delivers better performance than several trajectory variants.•Removal of a large number of trajectories related to background noise.•We apply our technique on action datasets HMDB51, UCF50 and UCF101 containing largest number of classes till date.

论文关键词：Action recognition,Dense trajectories,Large scale classification,Fisher vector,Bag-of-Words,SVM

论文评审过程：Received 1 May 2014, Revised 13 May 2015, Accepted 29 June 2015, Available online 18 July 2015, Version of Record 25 August 2015.

论文官网地址：https://doi.org/10.1016/j.imavis.2015.06.009