Gesture recognition: A review focusing on sign language in a mobile context
作者：
Highlights：
• Most used techniques: skin detection, brute force comparison and SVM.
• Classification scores: 39%-99% (static gestures); 61.3%-97.4% (dynamic gestures with special hardware).
• Predominant environment consisted of simple background and controlled light.
• Static and dynamic gestures are not recognized by the same approach.
• Gestures recognized and datasets are too small for real-world scenarios.
摘要
•Most used techniques: skin detection, brute force comparison and SVM.•Classification scores: 39%-99% (static gestures); 61.3%-97.4% (dynamic gestures with special hardware).•Predominant environment consisted of simple background and controlled light.•Static and dynamic gestures are not recognized by the same approach.•Gestures recognized and datasets are too small for real-world scenarios.
论文关键词：Gesture recognition,Sign language,Mobile devices
论文评审过程：Received 4 September 2017, Revised 28 January 2018, Accepted 29 January 2018, Available online 31 January 2018, Version of Record 20 March 2018.
论文官网地址：https://doi.org/10.1016/j.eswa.2018.01.051
