Systematic review of 3D facial expression recognition methods
作者：
Highlights：
• Recognition rates for Fear and Sadness expressions still remain a challenge.
• A trend towards keypoint-independent methods has been observed.
• There is a preference for local scope regions of interest for face representation.
• Significant information on preprocessing techniques have been neglected.
• The diversity of the experimentation setups limits the fair comparison among studies.
摘要
•Recognition rates for Fear and Sadness expressions still remain a challenge.•A trend towards keypoint-independent methods has been observed.•There is a preference for local scope regions of interest for face representation.•Significant information on preprocessing techniques have been neglected.•The diversity of the experimentation setups limits the fair comparison among studies.
论文关键词：3D facial expression recognition,Systematic literature review,Preprocessing techniques,Face representation,Deep learning,Classification setup
论文评审过程：Received 25 February 2019, Revised 3 October 2019, Accepted 10 November 2019, Available online 11 November 2019, Version of Record 16 November 2019.
论文官网地址：https://doi.org/10.1016/j.patcog.2019.107108
