Integrated inference and learning of neural factors in structural support vector machines

作者：

Highlights：

• A novel structured prediction model is proposed and applied to image segmentation.

• SSVM factors are modeled by highly nonlinear functions through neural networks.

• Back-propagation and loss-augmented inference are integrated in subgradient descent.

• Segmentation benchmark accuracy results show benefits over standard SSVM methods.

摘要

Highlights•A novel structured prediction model is proposed and applied to image segmentation.•SSVM factors are modeled by highly nonlinear functions through neural networks.•Back-propagation and loss-augmented inference are integrated in subgradient descent.•Segmentation benchmark accuracy results show benefits over standard SSVM methods.

论文关键词：Structural support vector machine,Neural factors,Structured prediction,Neural networks,Image segmentation

论文评审过程：Received 30 August 2015, Revised 8 March 2016, Accepted 9 March 2016, Available online 17 March 2016, Version of Record 23 August 2016.

论文官网地址：https://doi.org/10.1016/j.patcog.2016.03.014