-
Notifications
You must be signed in to change notification settings - Fork 0
Expand file tree
/
Copy pathREADME
More file actions
123 lines (85 loc) · 4.52 KB
/
Copy pathREADME
File metadata and controls
123 lines (85 loc) · 4.52 KB
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
SPtP (Smart-Point-to-Polygon) - User guide
by Philipp Naumann and Marcus Soll
-------------------------------------------------------------------------------
Contents:
0. Introduction
1. Dependencies
2. Batch processing
3. Graphical user interface
4. Unit tests
5. Learning
-------------------------------------------------------------------------------
0. Introduction
The lack of certain types of geographic data prevents the development of
location-aware technologies in a number of important domains. One such type of
“unmapped” geographic data is space usage rules (SURs), which are defined as
geographically-bound activity restrictions (e.g. “no dogs”, “no smoking”, “no
fishing”, “no skateboarding”). We present a novel SUR mapping technique - SPtP
- that outperforms current state-of-the-art approaches by 30% for one of the
most important components of the SUR mapping pipeline: associating a point
observation of a SUR (e.g. a ’no smoking’ sign) with the corresponding polygon
in which the SUR applies (e.g. the nearby park or the entire campus on which
the sign is located). The accompanying paper published at ACM CHI 2016 can be
found online at: http://dx.doi.org/10.1145/2858036.2858053
Marcus Soll, Philipp Naumann, Johannes Schöning, Pavel Samsonov & Brent Hecht:
Helping Computers Understand Geographically-Bound Activity Restrictions.
CHI 2016: Proceedings of the International Conference on Human Factors in
Computing Systems, 2016.
The data sets used for evaluation can be found at:
https://github.com/Top-Ranger/CHI2016-SUR-datasets
-------------------------------------------------------------------------------
1. Dependencies
SPtP requires Python 3. The following libraries need to be installed:
pillow (https://pypi.python.org/pypi/Pillow/2.6.1)
shapely (https://pypi.python.org/pypi/Shapely/1.5.0)
On most systems these can be installed using pip, e.g.
pip install pillow
pip install shapely
-------------------------------------------------------------------------------
2. Batch processing
The batch processing script (run_batch.py) processes a text file containing
locations. A location has a name, GPS coordinates and any number of space
usage rules. The script is started by running run_batch.py, e.g.
python3 run_batch.py
By default "./input/surs.txt" is the path to the locations file, "./input/"
is the path to the folder containing images (and *.truth.kml files), results
files are stored in "./output/" and log files are stored in "./log/".
Please use the command line option "--help" to display detailed information on
available command line options, e.g.
python3 run_batch.py --help
-------------------------------------------------------------------------------
3. Graphical user interface
The application features a web-based GUI which can be used with any modern web
browser. Running run_server.py will start the GUI server with default options,
e.g.:
python3 run_server.py
By default the server is configured to listen on localhost:8080 using the
batch processing script's default paths for images, *.truth.kml and
results files.
More detailed information on available command line options may be displayed
by using the command line option "--help", e.g.
python3 run_server.py --help
The server itself is a basic HTTP server. The web interface has three basic
functions:
1) Display a location previously processed by the batch processing script.
2) Manually enter a location to be processed. Uploading an image is optional
and only possible if the browser supports the FileReader API.
3) Select layers (map tiles, buildings, polygons)
The information box in the upper right corner may be expanded by left-clicking
it.
-------------------------------------------------------------------------------
4. Unit tests
The application's unit tests may be performed by running run_test.py, e.g.
python3 run_tests.py
-------------------------------------------------------------------------------
5. Learning
The learning process allows the generation of factors which determine the
weight of each weak classifier. A default factors file is included in this
distribution (/data/factors.txt). The learning process uses input data at
"./learning/db/" for the learning process and input data at
"./learning/test/" for verification.
The learning process can be started by running run_learning.py, e.g.:
python3 run_learning.py
The parameter of the learning process can be controlled through command line
arguments. For more information run
python3 run_learning.py --help