TH1.R9: Semantic Learning for Image Analysis

Thu, 1 Oct, 12:00 - 14:00 (UTC)
Thu, 1 Oct, 20:00 - 22:00 China Standard Time (UTC +8)
Thu, 1 Oct, 14:00 - 16:00 Central Europe Summer Time (UTC +2)
Thu, 1 Oct, 05:00 - 07:00 Pacific Daylight Time (UTC -7)
Session Co-Chairs: Stefano Zorzi, TUGraz and Bohao Huang, Duke University
Session Managers: Shashank S. Joshil and Na Liu
Track: Data Analysis Methods (Optical, Multispectral,Hyperspectral, SAR)
12:00-12:10 (UTC)
20:00-20:10 (UTC +8)
14:00-14:10 (UTC +2)
05:00-05:10 (UTC -7)

TH1.R9.1: SELF-CONSTRUCTING GRAPH CONVOLUTIONAL NETWORKS FOR SEMANTIC LABELING

Qinghui Liu, Norwegian Computing Center, Norway; Michael Kampffmeyer, Robert Jenssen, UiT The Arctic University of Norway, Norway; Arnt-Børre Salberg, Norwegian Computing Center, Norway
12:10-12:20 (UTC)
20:10-20:20 (UTC +8)
14:10-14:20 (UTC +2)
05:10-05:20 (UTC -7)

TH1.R9.2: REGULARIZED BUILDING SEGMENTATION BY FRAME FIELD LEARNING

Nicolas Girard, Inria, France; Dmitriy Smirnov, Justin Solomon, Massachusetts Institute of Technology, United States; Yuliya Tarabalka, LuxCarta Technology, France
12:20-12:30 (UTC)
20:20-20:30 (UTC +8)
14:20-14:30 (UTC +2)
05:20-05:30 (UTC -7)

TH1.R9.3: LOOK AT THE BIG PICTURE: BUILDING AREA EXTRACTION WITH GLOBAL DENSITY MAP

Haowen Guo, Wuhan University, China; Tongyuan Zou, Space Star Technology Co., Ltd. (SST), China; Wensheng Cheng, Wen Yang, Guisong Xia, Wuhan University, China
12:30-12:40 (UTC)
20:30-20:40 (UTC +8)
14:30-14:40 (UTC +2)
05:30-05:40 (UTC -7)

TH1.R9.4: SEMANTIC SEGMENTATION REFINEMENT WITH DEEP EDGE SUPERPIXELS TO ENHANCE HISTORICAL LAND COVER

Rémi Ratajczak, Carlos Crispim-Junior, Laboratoire d'InfoRmatique en Image et Systèmes d'information, France; Béatrice Fervers, Centre Léon Bérard, France; Elodie Faure, Gustave Roussy, France; Laure Tougne, Laboratoire d'InfoRmatique en Image et Systèmes d'information, France
12:40-12:50 (UTC)
20:40-20:50 (UTC +8)
14:40-14:50 (UTC +2)
05:40-05:50 (UTC -7)

TH1.R9.5: A MODIFIED D-LINKNET WITH TRANSFER LEARNING FOR ROAD EXTRACTION FROM HIGH-RESOLUTION REMOTE SENSING

Yanan Zhang, Qiqi Zhu, China University of Geosciences, China; Yanfei Zhong, Wuhan University, China; Qingfeng Guan, China University of Geosciences, China; Liangpei Zhang, Deren Li, Wuhan University, China
12:50-13:00 (UTC)
20:50-21:00 (UTC +8)
14:50-15:00 (UTC +2)
05:50-06:00 (UTC -7)

TH1.R9.6: LEARNING DISCRIMINATIVE GLOBAL AND LOCAL FEATURES FOR BUILDING EXTRACTION FROM AERIAL IMAGES

Yue Liao, Hongyan Zhang, Guangyi Yang, Liangpei Zhang, Wuhan University, China
13:00-13:10 (UTC)
21:00-21:10 (UTC +8)
15:00-15:10 (UTC +2)
06:00-06:10 (UTC -7)

TH1.R9.7: DILATED RESIDUAL NETWORK BASED ON DUAL EXPECTATION MAXIMIZATION ATTENTION FOR SEMANTIC SEGMENTATION OF REMOTE SENSING IMAGES

Jiachao Liu, Xinyue Xiong, Jiaojiao Li, Chaoxiong Wu, Rui Song, Xidian University, China
13:10-13:20 (UTC)
21:10-21:20 (UTC +8)
15:10-15:20 (UTC +2)
06:10-06:20 (UTC -7)

TH1.R9.8: MAP-REPAIR: DEEP CADASTRE MAPS ALIGNMENT AND TEMPORAL INCONSISTENCIES FIX IN SATELLITE IMAGES

Stefano Zorzi, TUGraz, Austria; Ksenia Bittner, German Aerospace Center (DLR), Germany; Friedrich Fraundorfer, TUGraz, Austria
13:20-13:30 (UTC)
21:20-21:30 (UTC +8)
15:20-15:30 (UTC +2)
06:20-06:30 (UTC -7)

TH1.R9.9: SEMANTIC SEGMENTATION OF URBAN BUILDINGS FROM VHR REMOTELY SENSED IMAGERY USING ATTENTION-BASED CNN

Zhijie Zhang, Chuanrong Zhang, Weidong Li, University of Connecticut, United States
13:30-13:40 (UTC)
21:30-21:40 (UTC +8)
15:30-15:40 (UTC +2)
06:30-06:40 (UTC -7)

TH1.R9.10: SEMI2I: SEMANTICALLY CONSISTENT IMAGE-TO-IMAGE TRANSLATION FOR DOMAIN ADAPTATION OF REMOTE SENSING DATA

Onur Tasar, S L Happy, Yuliya Tarabalka, Pierre Alliez, INRIA, France
13:40-13:50 (UTC)
21:40-21:50 (UTC +8)
15:40-15:50 (UTC +2)
06:40-06:50 (UTC -7)

TH1.R9.11: Spatial Attention Network for Road Extraction

Ruonan Chen, Yuan Hu, Tong Wu, Ling Peng, Aerospace Information Research Institute, Chinese Academy of Sciences, China