A weakly supervised method for makeup-invariant face verification
作者：
Highlights：
• Propose a weakly supervised method for face verification robust to cosmetic changes.
• Free video contexts are used to pre-train the proposed deep learning framework.
• Many techniques are used in the network to prevent overfitting.
• A large scale video face dataset and a before–after makeup dataset are collected.
• Our method achieves state-of-the-art performance on a benchmark dataset.
摘要
Highlights•Propose a weakly supervised method for face verification robust to cosmetic changes.•Free video contexts are used to pre-train the proposed deep learning framework.•Many techniques are used in the network to prevent overfitting.•A large scale video face dataset and a before–after makeup dataset are collected.•Our method achieves state-of-the-art performance on a benchmark dataset.
论文关键词：Face verification,Makeup-invariant,Weakly supervised method,Video context,Triplet loss function
论文评审过程：Received 15 July 2016, Revised 6 January 2017, Accepted 7 January 2017, Available online 10 January 2017, Version of Record 12 March 2017.
论文官网地址：https://doi.org/10.1016/j.patcog.2017.01.011
