Building Instance Classification Using Street View Images

Hannes Taubenb\"ock; Jian Kang; Marco K\"orner; Xiao Xiang Zhu; Yuanyuan Wang

RECORD

Building Instance Classification Using Street View Images

Jian Kang, Marco Körner, Yuanyuan Wang, Hannes Taubenböck, Xiao Xiang Zhu

Computer Vision and Pattern Recognition · Submitted 25 Feb 2018

abstract compiled · 2026-09-23 · no language-model review

The authors' main claim In addition, the method was applied to generate building classification maps on both region and city scales of several cities in Canada and the US.

how it is stated · asserted 1.00 read the record →

Comment (0) Read on arXiv ↗

Cite ▾

arXiv ID1802.09026 (version 1)

Pith ID7CLVTTPH, Pith's permanent record of this paper and its review

CitationHannes Taubenb\"ock, Jian Kang, Marco K\"orner, Xiao Xiang Zhu, Yuanyuan Wang. "Building Instance Classification Using Street View Images." arXiv:1802.09026 (2018). https://pith.science/paper/7CLVTTPH

Claim paper

The compiled record

The paper's own sentences, arranged. Each label is drawn from a closed vocabulary; the number beside it is the probability the classifier gave that label. Numbers in the tables are the authors', checked by code. No sentence in this record was written by a model.

Open the compiled record

Compiled by classifier-1.13.0 via TypeSafe API, question bank abstract-p1-20260923, 2026-09-23: 1 request, 7,366 input tokens.

What the abstract itself states, as probabilities: claims first 0.06 · states limitation 0.25 · human participants 0.03 · controlled comparison 0.05 · randomized assignment 0.01 · states quantitative result 0.04 · compares with prior results 0.08 · mentions code or data availability 0.05.

The abstract, sentence by sentence

arXiv listing abstract v1. Results and the main claim are underlined; show every role.

Land-use classification based on spaceborne or aerial remote sensing images has been extensively studied over the past decades. Such classification is usually a patch-wise or pixel-wise labeling over the whole image. But for many applications, such as urban population density mapping or urban utility planning, a classification map based on individual buildings is much more informative. However, such semantic classification still poses some fundamental challenges, for example, how to retrieve fine boundaries of individual buildings. In this paper, we proposed a general framework for classifying the functionality of individual buildings. The proposed method is based on Convolutional Neural Networks (CNNs) which classify facade structures from street view images, such as Google StreetView, in addition to remote sensing images which usually only show roof structures. Geographic information was utilized to mask out individual buildings, and to associate the corresponding street view images. We created a benchmark dataset which was used for training and evaluating CNNs. In addition, the method was applied to generate building classification maps on both region and city scales of several cities in Canada and the US. Keywords: CNN, Building instance classification, Street view images, OpenStreetMap

background contribution method result interpretation main claim

What the classifier reports for each sentence. The number is the probability it gave the label.
#	sentence	role	the authors' own work	states the main conclusion	strength
s0	Land-use classification based on spaceborne or aerial remote sensing images has been extensively studied over the past decades.	background 1.00	0.03	0.01	asserted 1.00
s1	Such classification is usually a patch-wise or pixel-wise labeling over the whole image.	background 1.00	0.05	0.02	asserted 0.98
s2	But for many applications, such as urban population density mapping or urban utility planning, a classification map based on individual buildings is much more informative.	background 0.60	0.06	0.04	asserted 0.95
s3	However, such semantic classification still poses some fundamental challenges, for example, how to retrieve fine boundaries of individual buildings.	gap 0.85	0.05	0.02	asserted 0.92
s4	In this paper, we proposed a general framework for classifying the functionality of individual buildings.	contribution 0.99	0.95	0.20	asserted 1.00
s5	The proposed method is based on Convolutional Neural Networks (CNNs) which classify facade structures from street view images, such as Google StreetView, in addition to remote sensing images which usually only show roof structures.	method 0.98	0.79	0.14	asserted 1.00
s6	Geographic information was utilized to mask out individual buildings, and to associate the corresponding street view images.	method 1.00	0.88	0.03	asserted 1.00
s7	We created a benchmark dataset which was used for training and evaluating CNNs.	method 0.67	0.96	0.04	asserted 1.00
s8	In addition, the method was applied to generate building classification maps on both region and city scales of several cities in Canada and the US.	result 0.56	0.95	0.20	asserted 1.00
s9	Keywords: CNN, Building instance classification, Street view images, OpenStreetMap	contribution 0.53	0.15	0.02	unclear 0.98

What kind of paper the classifier reports this is: method or tool 1.00 · evidence from observation or experiment 0.75.

What was examined, and what was not

sources read

arXiv listing abstract v1

not examined

full text
figures and tables
references and citing sentences
relations to other papers

levels run

P abstract

not run

S sections
T body sentences
C citations
R cross-paper relations

Every quoted sentence on this record is the authors'. Every label is from Pith's closed vocabulary, with the probability shown beside it. Every number in the measurements table is the authors', checked by code. Record as JSON

The paper

Original abstract

Land-use classification based on spaceborne or aerial remote sensing images has been extensively studied over the past decades. Such classification is usually a patch-wise or pixel-wise labeling over the whole image. But for many applications, such as urban population density mapping or urban utility planning, a classification map based on individual buildings is much more informative. However, such semantic classification still poses some fundamental challenges, for example, how to retrieve fine boundaries of individual buildings. In this paper, we proposed a general framework for classifying the functionality of individual buildings. The proposed method is based on Convolutional Neural Networks (CNNs) which classify facade structures from street view images, such as Google StreetView, in addition to remote sensing images which usually only show roof structures. Geographic information was utilized to mask out individual buildings, and to associate the corresponding street view images. We created a benchmark dataset which was used for training and evaluating CNNs. In addition, the method was applied to generate building classification maps on both region and city scales of several cities in Canada and the US. Keywords: CNN, Building instance classification, Street view images, OpenStreetMap

Human review

How Pith checks papers → · Full audit of this paper →

How the literature uses it

What this paper does with the arXiv papers it cites

Each citing sentence is taken from the citing paper's LaTeX source and typed by Pith's classifier. A label appears at probability 0.8 or more (0.95 for support), a contest at 0.7 or more; the quoted words are the citing authors'. How each label was checked · JSON

What it cites · 1 sentence without a confident label

Uses the method or tool of 2 papers

For removing them, we employ the released VGG16 model [ref] trained on Places2 dataset [ref] to preliminarily screen the street view images, as this architecture has achieved the highest top-1 accuracy[<https://github.com/metalbubble/places365>].

Very Deep Convolutional Networks for Large-Scale Image Recognition · Outlier removal by pretrained CNN on Places2 dataset · 0.98
For removing them, we employ the released VGG16 model [ref] trained on Places2 dataset [ref] to preliminarily screen the street view images, as this architecture has achieved the highest top-1 accuracy[<https://github.com/metalbubble/places365>].

Places: An Image Database for Deep Scene Understanding · Outlier removal by pretrained CNN on Places2 dataset · 0.98

Cites as background 3 papers

To improve the composition-based inference of land-use classes, multiscale CNN-based approaches are developed in [ref] .

Adaptive Deep Pyramid Matching for Remote Sensing Scene Classification · Related work · 0.82
The classification of land cover from Earth Observation (EO) images in complex urban environments has been a focus in remote sensing over the past decades [ref] .

Using convolutional networks and satellite imagery to identify patterns in urban environments at a large scale · Introduction · 0.99
In order to better interpret land-uses of Synthetic Aperture Radar (SAR) images in urban areas, [ref] proposes a pseudo-siamese CNN for identifying corresponding patches in very-high-resolution (VHR) optical and SAR remote sensing imagery.

Identifying Corresponding Patches in SAR and Optical Images with a Pseudo-Siamese CNN · Related work · 0.87

Discussion

Know something about this paper? Leave a comment, or highlight any passage above to comment on it.