IEEE IGARSS 2021 || Brussels, Belgium || July 12–14, 2021 • Physical

WE1.O-3.4

A MULTI-TASK DEEP LEARNING FRAMEWORK FOR BUILDING FOOTPRINT SEGMENTATION

Burak Ekim, Elif Sertel, Istanbul Technical University, Turkey

Session:

Advanced Segmentation and Land Cover Methods for Optical Data

Track:

Data Analysis Methods (Optical, Multispectral,Hyperspectral, SAR)

Presentation Time:

Wed, 14 Jul, 09:15-09:30 (UTC)
Wed, 14 Jul, 11:15-11:30 Central Europe Summer Time (UTC +2)
Wed, 14 Jul, 17:15-17:30 China Standard Time (UTC +8)
Wed, 14 Jul, 05:15-05:30 Eastern Daylight Time (UTC -4)

Session Co-Chairs:

Qingyu Li, Technical University of Munich, German Aerospace Center and Luc Baudoux, Institut national de l'information géographique et forestière

Session Manager:

Wufan Zhao

Session WE1.O-3

WE1.O-3.1: CONTEXTUAL LAND-COVER MAP TRANSLATION WITH SEMANTIC SEGMENTATION

Luc Baudoux, Institut national de l'information géographique et forestière, France; Jordi Inglada, Centre d'Etudes Spatiales de la Biosphère (CESBIO), France; Clément Mallet, Institut national de l'information géographique et forestière, France

WE1.O-3.2: Semi-Supervised Land-Use Classification Using Weakly Labeled Remote Sensing Data

Rui Wang, Man-On Pun, Chinese University of Hong Kong, Shenzhen, China; Huiliang Yu, Shanghai CAS-NOVA Satellite Technology Company Limited, China

WE1.O-3.3: LAND COVER CLASSIFICATION FROM A MAPPING PERSPECTIVE: PIXELWISE SUPERVISION IN THE DEEP LEARNING ERA

Thorsten Wilhelm, Dominik Koßmann, TU Dortmund University, Germany

WE1.O-3.4: A MULTI-TASK DEEP LEARNING FRAMEWORK FOR BUILDING FOOTPRINT SEGMENTATION

Burak Ekim, Elif Sertel, Istanbul Technical University, Turkey

WE1.O-3.5: HRLINKNET: LinkNet WITH HIGH-RESOLUTION REPRESENTATION FOR HIGH-RESOLUTION SATELLITE IMAGERY

Muyu Wu, Zhen Shu, Jinming Zhang, Xiangyun Hu, Wuhan University, China

WE1.O-3.6: End-to-end semantic segmentation and boundary regularization of buildings from satellite imagery

Qingyu Li, Technical University of Munich (TUM) / German Aerospace Center (DLR), Germany; Stefano Zorzi, Graz University of Technology, Austria; Yilei Shi, Technical University of Munich, Germany; Friedrich Fraundorfer, Graz University of Technology, German Aerospace Center, Austria; Xiao Xiang Zhu, Technical University of Munich (TUM) / German Aerospace Center (DLR), Germany

A MULTI-TASK DEEP LEARNING FRAMEWORK FOR BUILDING FOOTPRINT SEGMENTATION

Quicklinks

Downloads