MULTI-LEVEL FUSION OF THE MULTI-RECEPTIVE FIELDS CONTEXTUAL NETWORKS AND DISPARITY NETWORK FOR PAIRWISE SEMANTIC STEREO

被引：12

作者：

Chen, Hongyu ^{[1
]}

Lin, Manhui ^{[1
]}

Zhang, Hongyan ^{[1
]}

Yang, Guangyi ^{[2
]}

Xia, Gui-Song ^{[1
]}

Zheng, Xianwei ^{[1
]}

Zhang, Liangpei ^{[1
]}

机构：

[1] Wuhan Univ, State Key Lab Informat Engn Surveying Mapping & S, Wuhan 430079, Hubei, Peoples R China

[2] Wuhan Univ, Sch Elect Informat, Wuhan 430079, Hubei, Peoples R China

来源：

2019 IEEE INTERNATIONAL GEOSCIENCE AND REMOTE SENSING SYMPOSIUM (IGARSS 2019) | 2019年

基金：

中国国家自然科学基金;

关键词：

Data fusion; deep learning; semantic segmentation; disparity estimation; feature fusion;

D O I：

10.1109/igarss.2019.8899306

中图分类号：

P [天文学、地球科学];

学科分类号：

07 ;

摘要：

In this paper, we propose a multi-level fusion framework to address the pairwise semantic stereo issue. For disparity estimation, we adopt the pyramid stereo matching network. For semantic segmentation, the single segmentation network is proposed with respect to the left image, along with the disparity fusion segmentation network for the combination of semantic features and disparity features. Specifically, the multi-receptive fusion block is designed and employed to fully extract and fuse the contextual information. Finally, the refined segmentation result is obtained via yet another fusion of the multi-model results. The proposed method achieved a mean intersection over union (mIoU) of 79.05%, an average endpoint error (EPE) of 1.3966, and an mIoU-3 of 77.75%, ranking first in the Pairwise Semantic Stereo Challenge of the 2019 IEEE GRSS Data Fusion Contest [1,2].

引用

页码：4967 / 4970

页数：4

共 8 条

[1]

[Anonymous], PROC CVPR IEEE

[2]

[Anonymous], 2019, IEEE GEOSCI REMOTE S, DOI DOI 10.1109/LGRS.2019.2894399

[3]

[Anonymous], 2011, IEEE T PATTERN ANAL, DOI DOI 10.1109/TPAMI.2010.57

[4] Pyramid Stereo Matching Network [J].

Chang, Jia-Ren ;

Chen, Yong-Sheng .

2018 IEEE/CVF CONFERENCE ON COMPUTER VISION AND PATTERN RECOGNITION (CVPR), 2018, :5410-5418

[5]

Chen L, 2018, SODA'18: PROCEEDINGS OF THE TWENTY-NINTH ANNUAL ACM-SIAM SYMPOSIUM ON DISCRETE ALGORITHMS, P2801

[6]

Hu J, 2018, PROC CVPR IEEE, P7132, DOI [10.1109/CVPR.2018.00745, 10.1109/TPAMI.2019.2913372]

[7] Large Kernel Matters - Improve Semantic Segmentation by Global Convolutional Network [J].

Peng, Chao ;

Zhang, Xiangyu ;

Yu, Gang ;

Luo, Guiming ;

Sun, Jian .

30TH IEEE CONFERENCE ON COMPUTER VISION AND PATTERN RECOGNITION (CVPR 2017), 2017, :1743-1751

[8] Smoothed Dilated Convolutions for Improved Dense Prediction [J].

Wang, Zhengyang ;

Ji, Shuiwang .

KDD'18: PROCEEDINGS OF THE 24TH ACM SIGKDD INTERNATIONAL CONFERENCE ON KNOWLEDGE DISCOVERY & DATA MINING, 2018, :2486-2495

← 1 →