Approximate string matching: A lightweight approach to recognize gestures with Kinect

作者：

Highlights：

• We proposed a lightweight approach to recognize gestures with Kinect, based on approximate string matching.

• We implemented two gesture recognition variants, based on string matching.

• We evaluated our approach by using the public MSRC-12 Kinect gesture dataset.

• We compared the accuracy and performance with other state-of-the-art gesture-recognition techniques.

• The experimental evaluations show that the proposed approach achieves better performance than the state-of-the-art algorithms.

摘要

Highlights•We proposed a lightweight approach to recognize gestures with Kinect, based on approximate string matching.•We implemented two gesture recognition variants, based on string matching.•We evaluated our approach by using the public MSRC-12 Kinect gesture dataset.•We compared the accuracy and performance with other state-of-the-art gesture-recognition techniques.•The experimental evaluations show that the proposed approach achieves better performance than the state-of-the-art algorithms.

论文关键词：Natural user interfaces,Gesture recognition,Machine learning,Kinect,Approximate string matching

论文评审过程：Received 4 June 2015, Revised 14 July 2016, Accepted 20 August 2016, Available online 23 August 2016, Version of Record 7 September 2016.

论文官网地址：https://doi.org/10.1016/j.patcog.2016.08.022