Online view-invariant human action recognition using rgb-d spatio-temporal matrix

作者：

Highlights：

• We provide a promising vision-based view-invariant action recognition system from RGB-D videos.

• We use the Euclidean distance between spatio-temporal feature vectors that are represented in a Spatio-Temporal Matrix (STM).

• We describe the local tendency of the STM using pyramid-structural bag-of-words (BoW-Pyramid) and train a SVM classifier.

摘要

Highlights•We provide a promising vision-based view-invariant action recognition system from RGB-D videos.•We use the Euclidean distance between spatio-temporal feature vectors that are represented in a Spatio-Temporal Matrix (STM).•We describe the local tendency of the STM using pyramid-structural bag-of-words (BoW-Pyramid) and train a SVM classifier.

论文关键词：Action recognition,View-invariant,Self-similarity

论文评审过程：Received 10 June 2015, Revised 2 April 2016, Accepted 3 May 2016, Available online 26 May 2016, Version of Record 4 June 2016.

论文官网地址：https://doi.org/10.1016/j.patcog.2016.05.010