Computer vision uncovers predictors of physical urban change

被引:191
作者
Naik, Nikhil [1 ,2 ]
Kominers, Scott Duke [3 ,4 ,5 ,6 ,7 ,8 ]
Raskar, Ramesh [2 ]
Glaeser, Edward L. [5 ,8 ,9 ]
Hidalgo, Cesar A. [10 ]
机构
[1] Harvard Univ, Joint Ctr Hist & Econ, Cambridge, MA 02138 USA
[2] MIT, Media Lab, Cambridge, MA 02139 USA
[3] Harvard Univ, Soc Fellows, Cambridge, MA 02138 USA
[4] Harvard Sch Business, Boston, MA 02163 USA
[5] Harvard Univ, Dept Econ, Cambridge, MA 02138 USA
[6] Harvard Univ, Ctr Math Sci & Applicat, Cambridge, MA 02138 USA
[7] Harvard Univ, Ctr Res Computat & Soc, Cambridge, MA 02138 USA
[8] Natl Bur Econ Res, Cambridge, MA 02138 USA
[9] Harvard Univ, John F Kennedy Sch Govt, Cambridge, MA 02138 USA
[10] MIT, Media Lab, Collect Learning Grp, Cambridge, MA 02139 USA
基金
美国国家科学基金会;
关键词
urban economics; gentrification; urban studies; computer vision; neighborhood effects; GOOGLE STREET VIEW;
D O I
10.1073/pnas.1619003114
中图分类号
O [数理科学和化学]; P [天文学、地球科学]; Q [生物科学]; N [自然科学总论];
学科分类号
07 ; 0710 ; 09 ;
摘要
Which neighborhoods experience physical improvements? In this paper, weintroduce a computer vision method to measure changes in the physical appearances of neighborhoods from time-series street-level imagery. We connect changes in the physical appearance of five US cities with economic and demographic data and find three factors that predict neighborhood improvement. First, neighborhoods that are densely populated by college-educated adults are more likely to experience physical improvements-an observation that is compatible with the economic literature linking human capital and local success. Second, neighborhoods with better initial appearances experience, on average, larger positive improvements-an observation that is consistent with "tipping" theories of urban change. Third, neighborhood improvement correlates positively with physical proximity to the central business district and to other physically attractive neighborhoods-an observation that is consistent with the "invasion" theories of urban sociology. Together, our results provide support for three classical theories of urban change and illustrate the value of using computer vision methods and street-level imagery to understand the physical dynamics of cities.
引用
收藏
页码:7571 / 7576
页数:6
相关论文
共 37 条
[1]   GOOGLE STREET VIEW: CAPTURING THE WORLD AT STREET LEVEL [J].
Anguelov, Dragomir ;
Dulong, Carole ;
Filip, Daniel ;
Frueh, Christian ;
Lafon, Stephane ;
Lyon, Richard ;
Ogale, Abhijit ;
Vincent, Luc ;
Weaver, Josh .
COMPUTER, 2010, 43 (06) :32-38
[2]  
[Anonymous], 2015, CITIES
[3]  
[Anonymous], 1998, EVALUATIVE IMAGE CIT
[4]   City Forensics: Using Visual Elements to Predict Non-Visual City Attributes [J].
Arietta, Sean M. ;
Efros, Alexei A. ;
Ramamoorthi, Ravi ;
Agrawala, Maneesh .
IEEE TRANSACTIONS ON VISUALIZATION AND COMPUTER GRAPHICS, 2014, 20 (12) :2624-2633
[5]   Preserving history or restricting development? The heterogeneous effects of historic districts on local housing markets in New York City [J].
Been, Vicki ;
Ellen, Ingrid Gould ;
Gedal, Michael ;
Glaeser, Edward ;
McCabe, Brian J. .
JOURNAL OF URBAN ECONOMICS, 2016, 92 :16-30
[6]   The Origins of Scaling in Cities [J].
Bettencourt, Luis M. A. .
SCIENCE, 2013, 340 (6139) :1438-1441
[7]  
Burgess E., 1925, GROWTH CITY
[8]  
Ciccone A, 1996, AM ECON REV, V86, P54
[9]  
Conley TimothyG., 2008, The New Palgrave Dictionary of Economics, VSecond
[10]  
Cortright Joseph., 2014, Neighborhood Change 1970-2010: Transition and Growth in Urban High Poverty Neighborhoods