Towards Large-Scale Visual Mapping and Localization View Full Text


Ontology type: schema:Chapter     


Chapter Info

DATE

2011

AUTHORS

Marc Pollefeys , Jan-Michael Frahm , Friedrich Fraundorfer , Christopher Zach , Changchang Wu , Brian Clipp , David Gallup

ABSTRACT

The topic of this paper is large-scale mapping and localization from images. We first describe recent progress in obtaining large-scale 3D visual maps from images only. Our approach consists of a multi-stage processing pipeline, which can process a recorded video stream in real-time on standard PC hardware by leveraging the computational power of the graphics processor. The output of this pipeline is a detailed textured 3D model of the recorded area. The approach is demonstrated on video data recorded in Chapel Hill containing more than a million frames. While for these results GPS and inertial sensor data was used, we further explore the possibility to extract the necessary information for consistent 3D mapping over larger areas from images only. In particular, we discuss our recent work focusing on estimating the absolute scale of motion from images as well as finding intersections where the camera path crosses itself to effectively close loops in the mapping process. For this purpose we introduce viewpoint-invariant patches (VIP) as a new 3D feature that we extract from 3D models locally computed from the video sequence. These 3D features have important advantages with respect to traditional 2D SIFT features such as much stronger viewpoint-invariance, a relative pose hypothesis from a single match and a hierarchical matching scheme robust to repetitive structures. In addition, we also briefly discuss some additional work related to absolute scale estimation and multi-camera calibration. More... »

PAGES

535-555

Identifiers

URI

http://scigraph.springernature.com/pub.10.1007/978-3-642-19457-3_32

DOI

http://dx.doi.org/10.1007/978-3-642-19457-3_32

DIMENSIONS

https://app.dimensions.ai/details/publication/pub.1032571271


Indexing Status Check whether this publication has been indexed by Scopus and Web Of Science using the SN Indexing Status Tool
Incoming Citations Browse incoming citations for this publication using opencitations.net

JSON-LD is the canonical representation for SciGraph data.

TIP: You can open this SciGraph record using an external JSON-LD service: JSON-LD Playground Google SDTT

[
  {
    "@context": "https://springernature.github.io/scigraph/jsonld/sgcontext.json", 
    "about": [
      {
        "id": "http://purl.org/au-research/vocabulary/anzsrc-for/2008/08", 
        "inDefinedTermSet": "http://purl.org/au-research/vocabulary/anzsrc-for/2008/", 
        "name": "Information and Computing Sciences", 
        "type": "DefinedTerm"
      }, 
      {
        "id": "http://purl.org/au-research/vocabulary/anzsrc-for/2008/0801", 
        "inDefinedTermSet": "http://purl.org/au-research/vocabulary/anzsrc-for/2008/", 
        "name": "Artificial Intelligence and Image Processing", 
        "type": "DefinedTerm"
      }
    ], 
    "author": [
      {
        "affiliation": {
          "alternateName": "Institute of Visual Computing, ETH Z\u00fcrich", 
          "id": "http://www.grid.ac/institutes/None", 
          "name": [
            "Institute of Visual Computing, ETH Z\u00fcrich"
          ], 
          "type": "Organization"
        }, 
        "familyName": "Pollefeys", 
        "givenName": "Marc", 
        "id": "sg:person.013372156770.67", 
        "sameAs": [
          "https://app.dimensions.ai/discover/publication?and_facet_researcher=ur.013372156770.67"
        ], 
        "type": "Person"
      }, 
      {
        "affiliation": {
          "alternateName": "Dept. of Computer Science, University of North Carolina at Chapel Hill", 
          "id": "http://www.grid.ac/institutes/None", 
          "name": [
            "Dept. of Computer Science, University of North Carolina at Chapel Hill"
          ], 
          "type": "Organization"
        }, 
        "familyName": "Frahm", 
        "givenName": "Jan-Michael", 
        "id": "sg:person.01214540412.56", 
        "sameAs": [
          "https://app.dimensions.ai/discover/publication?and_facet_researcher=ur.01214540412.56"
        ], 
        "type": "Person"
      }, 
      {
        "affiliation": {
          "alternateName": "Institute of Visual Computing, ETH Z\u00fcrich", 
          "id": "http://www.grid.ac/institutes/None", 
          "name": [
            "Institute of Visual Computing, ETH Z\u00fcrich"
          ], 
          "type": "Organization"
        }, 
        "familyName": "Fraundorfer", 
        "givenName": "Friedrich", 
        "id": "sg:person.016452351373.65", 
        "sameAs": [
          "https://app.dimensions.ai/discover/publication?and_facet_researcher=ur.016452351373.65"
        ], 
        "type": "Person"
      }, 
      {
        "affiliation": {
          "alternateName": "Institute of Visual Computing, ETH Z\u00fcrich", 
          "id": "http://www.grid.ac/institutes/None", 
          "name": [
            "Institute of Visual Computing, ETH Z\u00fcrich"
          ], 
          "type": "Organization"
        }, 
        "familyName": "Zach", 
        "givenName": "Christopher", 
        "id": "sg:person.01240477356.10", 
        "sameAs": [
          "https://app.dimensions.ai/discover/publication?and_facet_researcher=ur.01240477356.10"
        ], 
        "type": "Person"
      }, 
      {
        "affiliation": {
          "alternateName": "Dept. of Computer Science, University of North Carolina at Chapel Hill", 
          "id": "http://www.grid.ac/institutes/None", 
          "name": [
            "Dept. of Computer Science, University of North Carolina at Chapel Hill"
          ], 
          "type": "Organization"
        }, 
        "familyName": "Wu", 
        "givenName": "Changchang", 
        "id": "sg:person.014706435321.29", 
        "sameAs": [
          "https://app.dimensions.ai/discover/publication?and_facet_researcher=ur.014706435321.29"
        ], 
        "type": "Person"
      }, 
      {
        "affiliation": {
          "alternateName": "Dept. of Computer Science, University of North Carolina at Chapel Hill", 
          "id": "http://www.grid.ac/institutes/None", 
          "name": [
            "Dept. of Computer Science, University of North Carolina at Chapel Hill"
          ], 
          "type": "Organization"
        }, 
        "familyName": "Clipp", 
        "givenName": "Brian", 
        "id": "sg:person.016301376321.34", 
        "sameAs": [
          "https://app.dimensions.ai/discover/publication?and_facet_researcher=ur.016301376321.34"
        ], 
        "type": "Person"
      }, 
      {
        "affiliation": {
          "alternateName": "Dept. of Computer Science, University of North Carolina at Chapel Hill", 
          "id": "http://www.grid.ac/institutes/None", 
          "name": [
            "Dept. of Computer Science, University of North Carolina at Chapel Hill"
          ], 
          "type": "Organization"
        }, 
        "familyName": "Gallup", 
        "givenName": "David", 
        "id": "sg:person.011062507733.99", 
        "sameAs": [
          "https://app.dimensions.ai/discover/publication?and_facet_researcher=ur.011062507733.99"
        ], 
        "type": "Person"
      }
    ], 
    "datePublished": "2011", 
    "datePublishedReg": "2011-01-01", 
    "description": "The topic of this paper is large-scale mapping and localization from images. We first describe recent progress in obtaining large-scale 3D visual maps from images only. Our approach consists of a multi-stage processing pipeline, which can process a recorded video stream in real-time on standard PC hardware by leveraging the computational power of the graphics processor. The output of this pipeline is a detailed textured 3D model of the recorded area. The approach is demonstrated on video data recorded in Chapel Hill containing more than a million frames. While for these results GPS and inertial sensor data was used, we further explore the possibility to extract the necessary information for consistent 3D mapping over larger areas from images only. In particular, we discuss our recent work focusing on estimating the absolute scale of motion from images as well as finding intersections where the camera path crosses itself to effectively close loops in the mapping process. For this purpose we introduce viewpoint-invariant patches (VIP) as a new 3D feature that we extract from 3D models locally computed from the video sequence. These 3D features have important advantages with respect to traditional 2D SIFT features such as much stronger viewpoint-invariance, a relative pose hypothesis from a single match and a hierarchical matching scheme robust to repetitive structures. In addition, we also briefly discuss some additional work related to absolute scale estimation and multi-camera calibration.", 
    "editor": [
      {
        "familyName": "Pradalier", 
        "givenName": "C\u00e9dric", 
        "type": "Person"
      }, 
      {
        "familyName": "Siegwart", 
        "givenName": "Roland", 
        "type": "Person"
      }, 
      {
        "familyName": "Hirzinger", 
        "givenName": "Gerhard", 
        "type": "Person"
      }
    ], 
    "genre": "chapter", 
    "id": "sg:pub.10.1007/978-3-642-19457-3_32", 
    "inLanguage": "en", 
    "isAccessibleForFree": false, 
    "isPartOf": {
      "isbn": [
        "978-3-642-19456-6", 
        "978-3-642-19457-3"
      ], 
      "name": "Robotics Research", 
      "type": "Book"
    }, 
    "keywords": [
      "multi-camera calibration", 
      "standard PC hardware", 
      "hierarchical matching scheme", 
      "new 3D features", 
      "absolute scale estimation", 
      "inertial sensor data", 
      "video streams", 
      "pose hypotheses", 
      "PC hardware", 
      "camera path", 
      "video sequences", 
      "video data", 
      "SIFT features", 
      "computational power", 
      "sensor data", 
      "graphics processors", 
      "processing pipeline", 
      "matching scheme", 
      "scale estimation", 
      "visual mapping", 
      "mapping process", 
      "necessary information", 
      "visual map", 
      "large-scale mapping", 
      "images", 
      "repetitive structure", 
      "pipeline", 
      "hardware", 
      "single match", 
      "processors", 
      "mapping", 
      "important advantages", 
      "features", 
      "scheme", 
      "recent work", 
      "GPS", 
      "information", 
      "work", 
      "streams", 
      "frame", 
      "path", 
      "estimation", 
      "model", 
      "match", 
      "maps", 
      "advantages", 
      "data", 
      "topic", 
      "localization", 
      "output", 
      "intersection", 
      "large areas", 
      "Chapel Hill", 
      "additional work", 
      "calibration", 
      "motion", 
      "power", 
      "absolute scale", 
      "area", 
      "process", 
      "sequence", 
      "loop", 
      "patches", 
      "purpose", 
      "progress", 
      "respect", 
      "recent progress", 
      "possibility", 
      "structure", 
      "addition", 
      "scale", 
      "Hill", 
      "hypothesis", 
      "approach", 
      "paper", 
      "large-scale 3D visual maps", 
      "multi-stage processing pipeline", 
      "results GPS", 
      "consistent 3D mapping", 
      "viewpoint-invariant patches", 
      "relative pose hypothesis", 
      "Large-Scale Visual Mapping"
    ], 
    "name": "Towards Large-Scale Visual Mapping and Localization", 
    "pagination": "535-555", 
    "productId": [
      {
        "name": "dimensions_id", 
        "type": "PropertyValue", 
        "value": [
          "pub.1032571271"
        ]
      }, 
      {
        "name": "doi", 
        "type": "PropertyValue", 
        "value": [
          "10.1007/978-3-642-19457-3_32"
        ]
      }
    ], 
    "publisher": {
      "name": "Springer Nature", 
      "type": "Organisation"
    }, 
    "sameAs": [
      "https://doi.org/10.1007/978-3-642-19457-3_32", 
      "https://app.dimensions.ai/details/publication/pub.1032571271"
    ], 
    "sdDataset": "chapters", 
    "sdDatePublished": "2021-12-01T20:04", 
    "sdLicense": "https://scigraph.springernature.com/explorer/license/", 
    "sdPublisher": {
      "name": "Springer Nature - SN SciGraph project", 
      "type": "Organization"
    }, 
    "sdSource": "s3://com-springernature-scigraph/baseset/20211201/entities/gbq_results/chapter/chapter_313.jsonl", 
    "type": "Chapter", 
    "url": "https://doi.org/10.1007/978-3-642-19457-3_32"
  }
]
 

Download the RDF metadata as:  json-ld nt turtle xml License info

HOW TO GET THIS DATA PROGRAMMATICALLY:

JSON-LD is a popular format for linked data which is fully compatible with JSON.

curl -H 'Accept: application/ld+json' 'https://scigraph.springernature.com/pub.10.1007/978-3-642-19457-3_32'

N-Triples is a line-based linked data format ideal for batch operations.

curl -H 'Accept: application/n-triples' 'https://scigraph.springernature.com/pub.10.1007/978-3-642-19457-3_32'

Turtle is a human-readable linked data format.

curl -H 'Accept: text/turtle' 'https://scigraph.springernature.com/pub.10.1007/978-3-642-19457-3_32'

RDF/XML is a standard XML format for linked data.

curl -H 'Accept: application/rdf+xml' 'https://scigraph.springernature.com/pub.10.1007/978-3-642-19457-3_32'


 

This table displays all metadata directly associated to this object as RDF triples.

196 TRIPLES      23 PREDICATES      108 URIs      101 LITERALS      7 BLANK NODES

Subject Predicate Object
1 sg:pub.10.1007/978-3-642-19457-3_32 schema:about anzsrc-for:08
2 anzsrc-for:0801
3 schema:author N6234e283a8f748909b010b9b38c6ea91
4 schema:datePublished 2011
5 schema:datePublishedReg 2011-01-01
6 schema:description The topic of this paper is large-scale mapping and localization from images. We first describe recent progress in obtaining large-scale 3D visual maps from images only. Our approach consists of a multi-stage processing pipeline, which can process a recorded video stream in real-time on standard PC hardware by leveraging the computational power of the graphics processor. The output of this pipeline is a detailed textured 3D model of the recorded area. The approach is demonstrated on video data recorded in Chapel Hill containing more than a million frames. While for these results GPS and inertial sensor data was used, we further explore the possibility to extract the necessary information for consistent 3D mapping over larger areas from images only. In particular, we discuss our recent work focusing on estimating the absolute scale of motion from images as well as finding intersections where the camera path crosses itself to effectively close loops in the mapping process. For this purpose we introduce viewpoint-invariant patches (VIP) as a new 3D feature that we extract from 3D models locally computed from the video sequence. These 3D features have important advantages with respect to traditional 2D SIFT features such as much stronger viewpoint-invariance, a relative pose hypothesis from a single match and a hierarchical matching scheme robust to repetitive structures. In addition, we also briefly discuss some additional work related to absolute scale estimation and multi-camera calibration.
7 schema:editor N26cc338920ef4dc9a21cb29dbb0b4a4a
8 schema:genre chapter
9 schema:inLanguage en
10 schema:isAccessibleForFree false
11 schema:isPartOf N4b08a9bdfe1645a2946b1da40742fa00
12 schema:keywords Chapel Hill
13 GPS
14 Hill
15 Large-Scale Visual Mapping
16 PC hardware
17 SIFT features
18 absolute scale
19 absolute scale estimation
20 addition
21 additional work
22 advantages
23 approach
24 area
25 calibration
26 camera path
27 computational power
28 consistent 3D mapping
29 data
30 estimation
31 features
32 frame
33 graphics processors
34 hardware
35 hierarchical matching scheme
36 hypothesis
37 images
38 important advantages
39 inertial sensor data
40 information
41 intersection
42 large areas
43 large-scale 3D visual maps
44 large-scale mapping
45 localization
46 loop
47 mapping
48 mapping process
49 maps
50 match
51 matching scheme
52 model
53 motion
54 multi-camera calibration
55 multi-stage processing pipeline
56 necessary information
57 new 3D features
58 output
59 paper
60 patches
61 path
62 pipeline
63 pose hypotheses
64 possibility
65 power
66 process
67 processing pipeline
68 processors
69 progress
70 purpose
71 recent progress
72 recent work
73 relative pose hypothesis
74 repetitive structure
75 respect
76 results GPS
77 scale
78 scale estimation
79 scheme
80 sensor data
81 sequence
82 single match
83 standard PC hardware
84 streams
85 structure
86 topic
87 video data
88 video sequences
89 video streams
90 viewpoint-invariant patches
91 visual map
92 visual mapping
93 work
94 schema:name Towards Large-Scale Visual Mapping and Localization
95 schema:pagination 535-555
96 schema:productId Na2bc295ecb614f61b28679d57442ba18
97 Na3d76dcb5f2847d1b99f35dfb3f969c2
98 schema:publisher N71e5f49422614ad1baea8dfdb61989f0
99 schema:sameAs https://app.dimensions.ai/details/publication/pub.1032571271
100 https://doi.org/10.1007/978-3-642-19457-3_32
101 schema:sdDatePublished 2021-12-01T20:04
102 schema:sdLicense https://scigraph.springernature.com/explorer/license/
103 schema:sdPublisher N0daac346c5bb44fbbf161d3514df383e
104 schema:url https://doi.org/10.1007/978-3-642-19457-3_32
105 sgo:license sg:explorer/license/
106 sgo:sdDataset chapters
107 rdf:type schema:Chapter
108 N07ce2260162149dd8cfaf490017230d8 rdf:first Nc53ddf2e1a77482786278b96c6a37c3d
109 rdf:rest N5213d866588b461c92a63902d0904f49
110 N0daac346c5bb44fbbf161d3514df383e schema:name Springer Nature - SN SciGraph project
111 rdf:type schema:Organization
112 N26cc338920ef4dc9a21cb29dbb0b4a4a rdf:first N4bb760080279447f8b5ce567a578fcb5
113 rdf:rest N07ce2260162149dd8cfaf490017230d8
114 N3a9683e65a8f43b587471e5cb98ddaad rdf:first sg:person.01240477356.10
115 rdf:rest N587006da056e4934a1800af286118c72
116 N44b978a0890d43c887fbd5b406e64107 rdf:first sg:person.016301376321.34
117 rdf:rest N48d838ab7e1c4d268729ae0e488b82a5
118 N48d838ab7e1c4d268729ae0e488b82a5 rdf:first sg:person.011062507733.99
119 rdf:rest rdf:nil
120 N4b08a9bdfe1645a2946b1da40742fa00 schema:isbn 978-3-642-19456-6
121 978-3-642-19457-3
122 schema:name Robotics Research
123 rdf:type schema:Book
124 N4bb760080279447f8b5ce567a578fcb5 schema:familyName Pradalier
125 schema:givenName Cédric
126 rdf:type schema:Person
127 N5213d866588b461c92a63902d0904f49 rdf:first Nc8546340d0ac4f89a3dc925e58a76a4b
128 rdf:rest rdf:nil
129 N587006da056e4934a1800af286118c72 rdf:first sg:person.014706435321.29
130 rdf:rest N44b978a0890d43c887fbd5b406e64107
131 N6234e283a8f748909b010b9b38c6ea91 rdf:first sg:person.013372156770.67
132 rdf:rest N96da1759a83d4c32b8c50f2fa92272c0
133 N71e5f49422614ad1baea8dfdb61989f0 schema:name Springer Nature
134 rdf:type schema:Organisation
135 N96da1759a83d4c32b8c50f2fa92272c0 rdf:first sg:person.01214540412.56
136 rdf:rest Nda25535cd8134505aa881ee6e49b8910
137 Na2bc295ecb614f61b28679d57442ba18 schema:name doi
138 schema:value 10.1007/978-3-642-19457-3_32
139 rdf:type schema:PropertyValue
140 Na3d76dcb5f2847d1b99f35dfb3f969c2 schema:name dimensions_id
141 schema:value pub.1032571271
142 rdf:type schema:PropertyValue
143 Nc53ddf2e1a77482786278b96c6a37c3d schema:familyName Siegwart
144 schema:givenName Roland
145 rdf:type schema:Person
146 Nc8546340d0ac4f89a3dc925e58a76a4b schema:familyName Hirzinger
147 schema:givenName Gerhard
148 rdf:type schema:Person
149 Nda25535cd8134505aa881ee6e49b8910 rdf:first sg:person.016452351373.65
150 rdf:rest N3a9683e65a8f43b587471e5cb98ddaad
151 anzsrc-for:08 schema:inDefinedTermSet anzsrc-for:
152 schema:name Information and Computing Sciences
153 rdf:type schema:DefinedTerm
154 anzsrc-for:0801 schema:inDefinedTermSet anzsrc-for:
155 schema:name Artificial Intelligence and Image Processing
156 rdf:type schema:DefinedTerm
157 sg:person.011062507733.99 schema:affiliation grid-institutes:None
158 schema:familyName Gallup
159 schema:givenName David
160 schema:sameAs https://app.dimensions.ai/discover/publication?and_facet_researcher=ur.011062507733.99
161 rdf:type schema:Person
162 sg:person.01214540412.56 schema:affiliation grid-institutes:None
163 schema:familyName Frahm
164 schema:givenName Jan-Michael
165 schema:sameAs https://app.dimensions.ai/discover/publication?and_facet_researcher=ur.01214540412.56
166 rdf:type schema:Person
167 sg:person.01240477356.10 schema:affiliation grid-institutes:None
168 schema:familyName Zach
169 schema:givenName Christopher
170 schema:sameAs https://app.dimensions.ai/discover/publication?and_facet_researcher=ur.01240477356.10
171 rdf:type schema:Person
172 sg:person.013372156770.67 schema:affiliation grid-institutes:None
173 schema:familyName Pollefeys
174 schema:givenName Marc
175 schema:sameAs https://app.dimensions.ai/discover/publication?and_facet_researcher=ur.013372156770.67
176 rdf:type schema:Person
177 sg:person.014706435321.29 schema:affiliation grid-institutes:None
178 schema:familyName Wu
179 schema:givenName Changchang
180 schema:sameAs https://app.dimensions.ai/discover/publication?and_facet_researcher=ur.014706435321.29
181 rdf:type schema:Person
182 sg:person.016301376321.34 schema:affiliation grid-institutes:None
183 schema:familyName Clipp
184 schema:givenName Brian
185 schema:sameAs https://app.dimensions.ai/discover/publication?and_facet_researcher=ur.016301376321.34
186 rdf:type schema:Person
187 sg:person.016452351373.65 schema:affiliation grid-institutes:None
188 schema:familyName Fraundorfer
189 schema:givenName Friedrich
190 schema:sameAs https://app.dimensions.ai/discover/publication?and_facet_researcher=ur.016452351373.65
191 rdf:type schema:Person
192 grid-institutes:None schema:alternateName Dept. of Computer Science, University of North Carolina at Chapel Hill
193 Institute of Visual Computing, ETH Zürich
194 schema:name Dept. of Computer Science, University of North Carolina at Chapel Hill
195 Institute of Visual Computing, ETH Zürich
196 rdf:type schema:Organization
 




Preview window. Press ESC to close (or click here)


...