Model compression of SDM-based face alignment for mobile applications

作     者：Shen, Yehu Jiang, Quansheng Yang, Yong Wang, Bangfu Zhu, Qixin 

作者机构：Suzhou Univ Sci & Technol Coll Mech Engn Suzhou Peoples R China Suzhou Key Lab Precis & Efficient Machining Techn Suzhou Peoples R China 

出 版 物：《JOURNAL OF ENGINEERING-JOE》 

年 卷 期：2018年第2018卷第16期

页      面：1712-1718页

学科分类：12[管理学] 1201[管理学-管理科学与工程(可授管理学、工学学位)] 08[工学] 

基　　金：National Natural Science Foundation of China [61501451, 51375323, 61563022] Cooperative Innovation Fund-Prospective of Jiangsu Province [BY2016044-01] Major Program of Natural Science Foundation of Jiangxi Province, China [20152ACB20009] high level talents of 'Six Talent Peaks' in Jiangsu Province, China [DZXX-046] Qing Lan Project of Jiangsu Province, China 

主　　题：image coding regression analysis learning (artificial intelligence) face recognition data compression quantisation (signal) gradient methods model compression SDM-based face alignment mobile applications face recognition expression recognition face-based AR applications cascaded-regression-based face alignment algorithms low computational costs impressive results trained model cascaded-regression-based methods commercial applications mobile phones data compression method supervised descent method model data nonparametric method adaptive quantisation algorithm SDM training process quantitative experimental results data model mobile AR application 

摘      要：Face alignment could be widely used in face recognition, expression recognition, face-based AR applications etc. Cascaded-regression-based face alignment algorithms have been popular in recent years for their low computational costs and impressive results in uncontrolled scenarios. Unfortunately, the size of the trained model is quite large for cascaded-regression-based methods which makes it unsuitable for commercial applications on mobile phones. In this study, the authors proposed a data compression method for the trained model of the supervised descent method (SDM). Firstly, the distribution of the model data was estimated using a non-parametric method. Then an adaptive quantisation algorithm was proposed to quantise the model data. Finally, their adaptive quantisation algorithm was tightly coupled with the SDM training process to fine tune the results. The quantitative experimental results proved that their proposed method could compress the data model to 20% of its original size without hurting the performances. The proposed method has been integrated into a mobile AR application, subjective evaluations proved that the proposed compression method could provide similar visual effects compared with the uncompressed counterpart.