Multitask Learning of Height and Semantics From Aerial Images

被引:43
作者
Carvalho, Marcela [1 ]
Le Saux, Bertrand [1 ]
Trouve-Peloux, Pauline [1 ]
Champagnat, Frederic [1 ]
Almansa, Andres [2 ]
机构
[1] Univ Paris Saclay, ONERA, DTIS, F-91123 Palaiseau, France
[2] Univ Paris 05, MAP5, F-75006 Paris, France
关键词
Semantics; Task analysis; Training; Estimation; Predictive models; Decoding; Land surface; Aerial imagery; deep learning; multitask learning; neural networks; semantic segmentation; single view depth estimation; DEEP; CLASSIFICATION;
D O I
10.1109/LGRS.2019.2947783
中图分类号
P3 [地球物理学]; P59 [地球化学];
学科分类号
0708 ; 070902 ;
摘要
Aerial or satellite imagery is a great source for land surface analysis, which might yield land-use maps or elevation models. In this letter, we present a neural network framework for learning semantics and local height together. We show how this joint multitask learning benefits to each task on the large data set of the 2018 Data Fusion Contest. Moreover, our framework also yields an uncertainty map that allows assessing the prediction of the model. Code is available at https://github.com/marcelampc/mtl_aerial_images
引用
收藏
页码:1391 / 1395
页数:5
相关论文
共 30 条
[1]   Height estimation from single aerial images using a deep convolutional encoder-decoder network [J].
Amirkolaee, Hamed Amini ;
Arefi, Hossein .
ISPRS JOURNAL OF PHOTOGRAMMETRY AND REMOTE SENSING, 2019, 149 :50-66
[2]  
[Anonymous], 2018, ISPRS J PHOTOGRAMM, DOI [DOI https://doi.org/10.1016/j.isprsjprs.2017.11.009, DOI 10.1016/j.isprsjprs.2017.11.009]
[3]  
[Anonymous], 2014, ResearcheGate, DOI DOI 10.13140/2.1.5015.9683
[4]   Beyond RGB: Very high resolution urban remote sensing with multimodal deep networks [J].
Audebert, Nicolas ;
Le Saux, Bertrand ;
Lefevre, Sebastien .
ISPRS JOURNAL OF PHOTOGRAMMETRY AND REMOTE SENSING, 2018, 140 :20-32
[5]   NEURAL NETWORK APPROACHES VERSUS STATISTICAL-METHODS IN CLASSIFICATION OF MULTISOURCE REMOTE-SENSING DATA [J].
BENEDIKTSSON, JA ;
SWAIN, PH ;
ERSOY, OK .
IEEE TRANSACTIONS ON GEOSCIENCE AND REMOTE SENSING, 1990, 28 (04) :540-552
[6]   Semantic object classes in video: A high-definition ground truth database [J].
Brostow, Gabriel J. ;
Fauqueur, Julien ;
Cipolla, Roberto .
PATTERN RECOGNITION LETTERS, 2009, 30 (02) :88-97
[7]   Multitask learning [J].
Caruana, R .
MACHINE LEARNING, 1997, 28 (01) :41-75
[8]  
Carvalho M, 2018, IEEE IMAGE PROC, P2915, DOI 10.1109/ICIP.2018.8451312
[9]  
Cerra D, 2018, INT GEOSCI REMOTE SE, P3856, DOI 10.1109/IGARSS.2018.8517699
[10]  
Chen Z, 2018, SPRINGER THESES-RECO, P1, DOI 10.1007/978-981-10-7703-6