Semi self-training beard/moustache detection and segmentation simultaneously

作者：

Highlights：

• Detect and segment beard/moustache simultaneously

• Use advantages of both pre-trained model and self-trained model

• Work on superpixel

• Propose an aggregate searching strategy to overcome the limits of landmarker

• Propose a new feature that is able to emphasize high frequency information of facial hair

摘要

•Detect and segment beard/moustache simultaneously•Use advantages of both pre-trained model and self-trained model•Work on superpixel•Propose an aggregate searching strategy to overcome the limits of landmarker•Propose a new feature that is able to emphasize high frequency information of facial hair

论文关键词：Beard/moustache,Detection and segmentation,Super pixel,Random Ferns,Support Vector Machine,Histogram of Gabor (HoG),Histogram of Oriented Gradient of Gabor (HOGG)

论文评审过程：Received 22 September 2015, Revised 15 July 2016, Accepted 25 July 2016, Available online 21 September 2016, Version of Record 20 February 2017.

论文官网地址：https://doi.org/10.1016/j.imavis.2016.07.009