Deep Hierarchical Parsing for Semantic Segmentation

被引:0
作者
Sharma, Abhishek [1 ]
Tuzel, Oncel [2 ]
Jacobs, David W. [1 ]
机构
[1] Univ Maryland, Dept Comp Sci, College Pk, MD 20742 USA
[2] MERL, Cambridge, MA USA
来源
2015 IEEE CONFERENCE ON COMPUTER VISION AND PATTERN RECOGNITION (CVPR) | 2015年
关键词
D O I
暂无
中图分类号
TP18 [人工智能理论];
学科分类号
081104 ; 0812 ; 0835 ; 1405 ;
摘要
This paper proposes a learning-based approach to scene parsing inspired by the deep Recursive Context Propagation Network (RCPN). RCPN is a deep feed-forward neural network that utilizes the contextual information from the entire image, through bottom-up followed by top-down context propagation via random binary parse trees. This improves the feature representation of every super-pixel in the image for better classification into semantic categories. We analyze RCPN and propose two novel contributions to further improve the model. We first analyze the learning of RCPN parameters and discover the presence of bypass error paths in the computation graph of RCPN that can hinder contextual propagation. We propose to tackle this problem by including the classification loss of the internal nodes of the random parse trees in the original RCPN loss function. Secondly, we use an MRF on the parse tree nodes to model the hierarchical dependency present in the output. Both modifications provide performance boosts over the original RCPN and the new system achieves state-of-the-art performance on Stanford Background, SIFT-Flow and Daimler urban datasets.
引用
收藏
页码:530 / 538
页数:9
相关论文
共 25 条
[1]  
[Anonymous], IEEE TPAMI
[2]  
[Anonymous], P 26 INT C MACH LEAR
[3]  
[Anonymous], ECCV
[4]  
[Anonymous], CVPR
[5]  
[Anonymous], 2013, Caffe: An Open Source Convolutional Architecture for Fast Feature Embedding
[6]  
[Anonymous], IEEE T PAMI
[7]  
[Anonymous], 2014, ICML
[8]  
Erhan D, 2010, J MACH LEARN RES, V11, P625
[9]   Efficient graph-based image segmentation [J].
Felzenszwalb, PF ;
Huttenlocher, DP .
INTERNATIONAL JOURNAL OF COMPUTER VISION, 2004, 59 (02) :167-181
[10]  
Fergus R., 2012, IEEE CVPR