Topometric Localization with Deep Learning View Full Text


Ontology type: schema:Chapter      Open Access: True


Chapter Info

DATE

2019-11-28

AUTHORS

Gabriel L. Oliveira , Noha Radwan , Wolfram Burgard , Thomas Brox

ABSTRACT

Compared to LiDAR-based localization methods, which provide high accuracy but rely on expensive sensors, visual localization approaches only require a camera and thus are more cost-effective however their accuracy and reliability is typically inferior to LiDAR-based methods. In this work, we propose a vision-based localization approach that learns from LiDAR-based localization methods by using their output as training data, thus combining a cheap, passive sensor with an accuracy that is on-par with LiDAR-based localization. The approach consists of two deep networks trained on visual odometry and topological localization, respectively, and a successive optimization to combine the predictions of these two networks. Furthermore, we introduce a new challenging pedestrian-based dataset for localization with a high degree of noise. Results obtained by evaluating the proposed approach on this novel dataset demonstrate localization errors up to 10 times smaller than those obtained with traditional vision-based localization methods. More... »

PAGES

505-520

Identifiers

URI

http://scigraph.springernature.com/pub.10.1007/978-3-030-28619-4_38

DOI

http://dx.doi.org/10.1007/978-3-030-28619-4_38

DIMENSIONS

https://app.dimensions.ai/details/publication/pub.1122954184


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": "Department of Computer Science, University of Freiburg, Freiburg im Breisgau, Germany", 
          "id": "http://www.grid.ac/institutes/grid.5963.9", 
          "name": [
            "Department of Computer Science, University of Freiburg, Freiburg im Breisgau, Germany"
          ], 
          "type": "Organization"
        }, 
        "familyName": "Oliveira", 
        "givenName": "Gabriel L.", 
        "id": "sg:person.011310276047.71", 
        "sameAs": [
          "https://app.dimensions.ai/discover/publication?and_facet_researcher=ur.011310276047.71"
        ], 
        "type": "Person"
      }, 
      {
        "affiliation": {
          "alternateName": "Department of Computer Science, University of Freiburg, Freiburg im Breisgau, Germany", 
          "id": "http://www.grid.ac/institutes/grid.5963.9", 
          "name": [
            "Department of Computer Science, University of Freiburg, Freiburg im Breisgau, Germany"
          ], 
          "type": "Organization"
        }, 
        "familyName": "Radwan", 
        "givenName": "Noha", 
        "id": "sg:person.011714173446.93", 
        "sameAs": [
          "https://app.dimensions.ai/discover/publication?and_facet_researcher=ur.011714173446.93"
        ], 
        "type": "Person"
      }, 
      {
        "affiliation": {
          "alternateName": "Department of Computer Science, University of Freiburg, Freiburg im Breisgau, Germany", 
          "id": "http://www.grid.ac/institutes/grid.5963.9", 
          "name": [
            "Department of Computer Science, University of Freiburg, Freiburg im Breisgau, Germany"
          ], 
          "type": "Organization"
        }, 
        "familyName": "Burgard", 
        "givenName": "Wolfram", 
        "id": "sg:person.014270043511.25", 
        "sameAs": [
          "https://app.dimensions.ai/discover/publication?and_facet_researcher=ur.014270043511.25"
        ], 
        "type": "Person"
      }, 
      {
        "affiliation": {
          "alternateName": "Department of Computer Science, University of Freiburg, Freiburg im Breisgau, Germany", 
          "id": "http://www.grid.ac/institutes/grid.5963.9", 
          "name": [
            "Department of Computer Science, University of Freiburg, Freiburg im Breisgau, Germany"
          ], 
          "type": "Organization"
        }, 
        "familyName": "Brox", 
        "givenName": "Thomas", 
        "id": "sg:person.012443225372.65", 
        "sameAs": [
          "https://app.dimensions.ai/discover/publication?and_facet_researcher=ur.012443225372.65"
        ], 
        "type": "Person"
      }
    ], 
    "datePublished": "2019-11-28", 
    "datePublishedReg": "2019-11-28", 
    "description": "Compared to LiDAR-based localization methods, which provide high accuracy but rely on expensive sensors, visual localization approaches only require a camera and thus are more cost-effective however their accuracy and reliability is typically inferior to LiDAR-based methods. In this work, we propose a vision-based localization approach that learns from LiDAR-based localization methods by using their output as training data, thus combining a cheap, passive sensor with an accuracy that is on-par with LiDAR-based localization. The approach consists of two deep networks trained on visual odometry and topological localization, respectively, and a successive optimization to combine the predictions of these two networks. Furthermore, we introduce a new challenging pedestrian-based dataset for localization with a high degree of noise. Results obtained by evaluating the proposed approach on this novel dataset demonstrate localization errors up\u00a0to 10 times smaller than those obtained with traditional vision-based localization methods.", 
    "editor": [
      {
        "familyName": "Amato", 
        "givenName": "Nancy M.", 
        "type": "Person"
      }, 
      {
        "familyName": "Hager", 
        "givenName": "Greg", 
        "type": "Person"
      }, 
      {
        "familyName": "Thomas", 
        "givenName": "Shawna", 
        "type": "Person"
      }, 
      {
        "familyName": "Torres-Torriti", 
        "givenName": "Miguel", 
        "type": "Person"
      }
    ], 
    "genre": "chapter", 
    "id": "sg:pub.10.1007/978-3-030-28619-4_38", 
    "isAccessibleForFree": true, 
    "isPartOf": {
      "isbn": [
        "978-3-030-28618-7", 
        "978-3-030-28619-4"
      ], 
      "name": "Robotics Research", 
      "type": "Book"
    }, 
    "keywords": [
      "localization method", 
      "vision-based localization approach", 
      "localization approach", 
      "vision-based localization method", 
      "visual localization approach", 
      "deep network", 
      "deep learning", 
      "visual odometry", 
      "training data", 
      "topological localization", 
      "expensive sensors", 
      "localization error", 
      "high accuracy", 
      "successive optimization", 
      "network", 
      "accuracy", 
      "passive sensors", 
      "lidar", 
      "odometry", 
      "sensors", 
      "camera", 
      "dataset", 
      "learning", 
      "method", 
      "optimization", 
      "localization", 
      "reliability", 
      "error", 
      "noise", 
      "output", 
      "work", 
      "prediction", 
      "high degree", 
      "data", 
      "par", 
      "time", 
      "results", 
      "degree", 
      "approach"
    ], 
    "name": "Topometric Localization with Deep Learning", 
    "pagination": "505-520", 
    "productId": [
      {
        "name": "dimensions_id", 
        "type": "PropertyValue", 
        "value": [
          "pub.1122954184"
        ]
      }, 
      {
        "name": "doi", 
        "type": "PropertyValue", 
        "value": [
          "10.1007/978-3-030-28619-4_38"
        ]
      }
    ], 
    "publisher": {
      "name": "Springer Nature", 
      "type": "Organisation"
    }, 
    "sameAs": [
      "https://doi.org/10.1007/978-3-030-28619-4_38", 
      "https://app.dimensions.ai/details/publication/pub.1122954184"
    ], 
    "sdDataset": "chapters", 
    "sdDatePublished": "2022-09-02T16:13", 
    "sdLicense": "https://scigraph.springernature.com/explorer/license/", 
    "sdPublisher": {
      "name": "Springer Nature - SN SciGraph project", 
      "type": "Organization"
    }, 
    "sdSource": "s3://com-springernature-scigraph/baseset/20220902/entities/gbq_results/chapter/chapter_22.jsonl", 
    "type": "Chapter", 
    "url": "https://doi.org/10.1007/978-3-030-28619-4_38"
  }
]
 

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.10.1007/978-3-030-28619-4_38'

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

curl -H 'Accept: application/n-triples' 'https://scigraph.springernature.com/pub.10.1007/978-3-030-28619-4_38'

Turtle is a human-readable linked data format.

curl -H 'Accept: text/turtle' 'https://scigraph.springernature.com/pub.10.1007/978-3-030-28619-4_38'

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

curl -H 'Accept: application/rdf+xml' 'https://scigraph.springernature.com/pub.10.1007/978-3-030-28619-4_38'


 

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

134 TRIPLES      22 PREDICATES      63 URIs      56 LITERALS      7 BLANK NODES

Subject Predicate Object
1 sg:pub.10.1007/978-3-030-28619-4_38 schema:about anzsrc-for:08
2 anzsrc-for:0801
3 schema:author N82ad0f69076d4f8dafac7ec27394374e
4 schema:datePublished 2019-11-28
5 schema:datePublishedReg 2019-11-28
6 schema:description Compared to LiDAR-based localization methods, which provide high accuracy but rely on expensive sensors, visual localization approaches only require a camera and thus are more cost-effective however their accuracy and reliability is typically inferior to LiDAR-based methods. In this work, we propose a vision-based localization approach that learns from LiDAR-based localization methods by using their output as training data, thus combining a cheap, passive sensor with an accuracy that is on-par with LiDAR-based localization. The approach consists of two deep networks trained on visual odometry and topological localization, respectively, and a successive optimization to combine the predictions of these two networks. Furthermore, we introduce a new challenging pedestrian-based dataset for localization with a high degree of noise. Results obtained by evaluating the proposed approach on this novel dataset demonstrate localization errors up to 10 times smaller than those obtained with traditional vision-based localization methods.
7 schema:editor N34a1dc37bd224ea0a322d6473cd86882
8 schema:genre chapter
9 schema:isAccessibleForFree true
10 schema:isPartOf Nf63de548b4aa419fbf193458156a8404
11 schema:keywords accuracy
12 approach
13 camera
14 data
15 dataset
16 deep learning
17 deep network
18 degree
19 error
20 expensive sensors
21 high accuracy
22 high degree
23 learning
24 lidar
25 localization
26 localization approach
27 localization error
28 localization method
29 method
30 network
31 noise
32 odometry
33 optimization
34 output
35 par
36 passive sensors
37 prediction
38 reliability
39 results
40 sensors
41 successive optimization
42 time
43 topological localization
44 training data
45 vision-based localization approach
46 vision-based localization method
47 visual localization approach
48 visual odometry
49 work
50 schema:name Topometric Localization with Deep Learning
51 schema:pagination 505-520
52 schema:productId N3dab75af6e60450394c571bd37975c73
53 Nb3abcb8fc8e9430488d832fc0c083cdb
54 schema:publisher Na32b2e00622c4c308f2c616b0ab48d40
55 schema:sameAs https://app.dimensions.ai/details/publication/pub.1122954184
56 https://doi.org/10.1007/978-3-030-28619-4_38
57 schema:sdDatePublished 2022-09-02T16:13
58 schema:sdLicense https://scigraph.springernature.com/explorer/license/
59 schema:sdPublisher N7a4e15f6fd9546958359d7923eb20ce6
60 schema:url https://doi.org/10.1007/978-3-030-28619-4_38
61 sgo:license sg:explorer/license/
62 sgo:sdDataset chapters
63 rdf:type schema:Chapter
64 N13223172a73f4c0e85048eee8bed4b1e rdf:first Nfd56d7ca24764f8387d22a9e4bf5a2c8
65 rdf:rest N3a8da2e1d1fb44f9ab37b1d99454f3f3
66 N1eab0d10aed14a4dbb1ee6dfae153869 schema:familyName Hager
67 schema:givenName Greg
68 rdf:type schema:Person
69 N34a1dc37bd224ea0a322d6473cd86882 rdf:first Nd81c3f7d84804bbc9ade8eb73bb852f0
70 rdf:rest N9e9212daeda5433e83b352ef40da91cc
71 N3a8da2e1d1fb44f9ab37b1d99454f3f3 rdf:first Nc149cb7f79624274b5304d21ab91b903
72 rdf:rest rdf:nil
73 N3dab75af6e60450394c571bd37975c73 schema:name doi
74 schema:value 10.1007/978-3-030-28619-4_38
75 rdf:type schema:PropertyValue
76 N4d6e1707b5654dc0af8ec8e1396e9ce1 rdf:first sg:person.011714173446.93
77 rdf:rest N52db50374e8641f5aad002ef82459a1c
78 N52db50374e8641f5aad002ef82459a1c rdf:first sg:person.014270043511.25
79 rdf:rest Naec9e4a2459f460083f768d245d2c233
80 N7a4e15f6fd9546958359d7923eb20ce6 schema:name Springer Nature - SN SciGraph project
81 rdf:type schema:Organization
82 N82ad0f69076d4f8dafac7ec27394374e rdf:first sg:person.011310276047.71
83 rdf:rest N4d6e1707b5654dc0af8ec8e1396e9ce1
84 N9e9212daeda5433e83b352ef40da91cc rdf:first N1eab0d10aed14a4dbb1ee6dfae153869
85 rdf:rest N13223172a73f4c0e85048eee8bed4b1e
86 Na32b2e00622c4c308f2c616b0ab48d40 schema:name Springer Nature
87 rdf:type schema:Organisation
88 Naec9e4a2459f460083f768d245d2c233 rdf:first sg:person.012443225372.65
89 rdf:rest rdf:nil
90 Nb3abcb8fc8e9430488d832fc0c083cdb schema:name dimensions_id
91 schema:value pub.1122954184
92 rdf:type schema:PropertyValue
93 Nc149cb7f79624274b5304d21ab91b903 schema:familyName Torres-Torriti
94 schema:givenName Miguel
95 rdf:type schema:Person
96 Nd81c3f7d84804bbc9ade8eb73bb852f0 schema:familyName Amato
97 schema:givenName Nancy M.
98 rdf:type schema:Person
99 Nf63de548b4aa419fbf193458156a8404 schema:isbn 978-3-030-28618-7
100 978-3-030-28619-4
101 schema:name Robotics Research
102 rdf:type schema:Book
103 Nfd56d7ca24764f8387d22a9e4bf5a2c8 schema:familyName Thomas
104 schema:givenName Shawna
105 rdf:type schema:Person
106 anzsrc-for:08 schema:inDefinedTermSet anzsrc-for:
107 schema:name Information and Computing Sciences
108 rdf:type schema:DefinedTerm
109 anzsrc-for:0801 schema:inDefinedTermSet anzsrc-for:
110 schema:name Artificial Intelligence and Image Processing
111 rdf:type schema:DefinedTerm
112 sg:person.011310276047.71 schema:affiliation grid-institutes:grid.5963.9
113 schema:familyName Oliveira
114 schema:givenName Gabriel L.
115 schema:sameAs https://app.dimensions.ai/discover/publication?and_facet_researcher=ur.011310276047.71
116 rdf:type schema:Person
117 sg:person.011714173446.93 schema:affiliation grid-institutes:grid.5963.9
118 schema:familyName Radwan
119 schema:givenName Noha
120 schema:sameAs https://app.dimensions.ai/discover/publication?and_facet_researcher=ur.011714173446.93
121 rdf:type schema:Person
122 sg:person.012443225372.65 schema:affiliation grid-institutes:grid.5963.9
123 schema:familyName Brox
124 schema:givenName Thomas
125 schema:sameAs https://app.dimensions.ai/discover/publication?and_facet_researcher=ur.012443225372.65
126 rdf:type schema:Person
127 sg:person.014270043511.25 schema:affiliation grid-institutes:grid.5963.9
128 schema:familyName Burgard
129 schema:givenName Wolfram
130 schema:sameAs https://app.dimensions.ai/discover/publication?and_facet_researcher=ur.014270043511.25
131 rdf:type schema:Person
132 grid-institutes:grid.5963.9 schema:alternateName Department of Computer Science, University of Freiburg, Freiburg im Breisgau, Germany
133 schema:name Department of Computer Science, University of Freiburg, Freiburg im Breisgau, Germany
134 rdf:type schema:Organization
 




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


...