Residual multi-task learning for facial landmark localization and expression recognition

作者：

Highlights：

• A novel residual learning framework for multi-task.

• An association learning method for multi-task learning.

• Train multi-task network with single type annotation.

• Yield superior performance in four benchmarks.

摘要

•A novel residual learning framework for multi-task.•An association learning method for multi-task learning.•Train multi-task network with single type annotation.•Yield superior performance in four benchmarks.

论文关键词：Facial landmark localization,Facial expression recognition,Deep neural network,Multi-task learning

论文评审过程：Received 7 January 2020, Revised 22 January 2021, Accepted 11 February 2021, Available online 20 February 2021, Version of Record 26 February 2021.

论文官网地址：https://doi.org/10.1016/j.patcog.2021.107893