Scene parsing using inference Embedded Deep Networks

作者：

Highlights：

• We design a novel structure of networks considering CRFs model as one type layer of deep neural networks.

• CRF is regarded as a layer of the network, therefore, the structural learning can be conducted explicitly.

• A novel feature encoding spatial relationship between objects in images is proposed.

• Feature fusing is adopted to learn intrinsic non-linear relationships between hierarchical and spatial features.

摘要

Highlights•We design a novel structure of networks considering CRFs model as one type layer of deep neural networks.•CRF is regarded as a layer of the network, therefore, the structural learning can be conducted explicitly.•A novel feature encoding spatial relationship between objects in images is proposed.•Feature fusing is adopted to learn intrinsic non-linear relationships between hierarchical and spatial features.

论文关键词：Convolutional Neural Networks (CNNs),Conditional Random Fields (CRFs),Inference Embedded Deep Networks (IEDNs),Hybrid Features

论文评审过程：Received 5 August 2015, Revised 10 December 2015, Accepted 22 January 2016, Available online 4 February 2016, Version of Record 23 August 2016.

论文官网地址：https://doi.org/10.1016/j.patcog.2016.01.027