Functional Map of the World

被引:175
作者
Christie, Gordon [1 ]
Fendley, Neil [1 ]
Wilson, James [2 ]
Mukherjee, Ryan [1 ]
机构
[1] Johns Hopkins Univ, Appl Phys Lab, Baltimore, MD 21218 USA
[2] DigitalGlobe, Westminster, CO USA
来源
2018 IEEE/CVF CONFERENCE ON COMPUTER VISION AND PATTERN RECOGNITION (CVPR) | 2018年
关键词
D O I
10.1109/CVPR.2018.00646
中图分类号
TP18 [人工智能理论];
学科分类号
081104 ; 0812 ; 0835 ; 1405 ;
摘要
We present a new dataset, Functional Map of the World (fMoW), which aims to inspire the development of machine learning models capable of predicting the functional purpose of buildings and land use from temporal sequences of satellite images and a rich set of metadata features. The metadata provided with each image enables reasoning about location, time, sun angles, physical sizes, and other features when making predictions about objects in the image. Our dataset consists of over 1 million images from over 200 countries1. For each image, we provide at least one bounding box annotation containing one of 63 categories, including a "false detection" category. We present an analysis of the dataset along with baseline approaches that reason about metadata and temporal views. Our data, code, and pretrained models have been made publicly available.
引用
收藏
页码:6172 / 6180
页数:9
相关论文
共 31 条
[1]  
[Anonymous], 2010, ACM GIS
[2]  
[Anonymous], 2014, ECCV
[3]  
[Anonymous], 2017, OPEN IMAGES PUBLIC D
[4]  
[Anonymous], 2016, arXiv
[5]  
[Anonymous], ASPRS
[6]  
[Anonymous], SCIENCE
[7]  
[Anonymous], 2016, EUR C COMP VIS
[8]  
[Anonymous], COSMIQ WORKS SPACENE
[9]  
[Anonymous], J STARS
[10]  
[Anonymous], CVPR