Learning to Select Long-Track Features for Structure-From-Motion and Visual SLAM View Full Text


Ontology type: schema:Chapter     


Chapter Info

DATE

2016-08-27

AUTHORS

Jonas Scheer , Mario Fritz , Oliver Grau

ABSTRACT

With the emergence of augmented reality platforms, Structure-From-Motion or visual SLAM approaches have regained in importance in order to deliver the next generation of immersive 3D experiences. As a new quality is achieved by deployment on mobile devices, computational efficiency plays an important role. In this work, we aim to reduce complexity by limiting the number of features without sacrificing quality. We select a subset of image features, using a learning based approach. A random forest is trained to pick 2D image features which are likely to be significant for a 3D reconstruction. Additionally, we aim for an objective that selects long track features, so that they can be “re-used” in multiple frames. We evaluate our feature selection technique on real world sequences and show a significant reduction of image features and the resulting decreased computation time is not effecting the accuracy of the 3D reconstruction. More... »

PAGES

402-413

Identifiers

URI

http://scigraph.springernature.com/pub.10.1007/978-3-319-45886-1_33

DOI

http://dx.doi.org/10.1007/978-3-319-45886-1_33

DIMENSIONS

https://app.dimensions.ai/details/publication/pub.1019686752


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": "Intel Visual Computing Institute, Saarbr\u00fccken, Germany", 
          "id": "http://www.grid.ac/institutes/None", 
          "name": [
            "Intel Visual Computing Institute, Saarbr\u00fccken, Germany"
          ], 
          "type": "Organization"
        }, 
        "familyName": "Scheer", 
        "givenName": "Jonas", 
        "type": "Person"
      }, 
      {
        "affiliation": {
          "alternateName": "Max-Planck Institute for Informatics, Saarbr\u00fccken, Germany", 
          "id": "http://www.grid.ac/institutes/grid.419528.3", 
          "name": [
            "Max-Planck Institute for Informatics, Saarbr\u00fccken, Germany"
          ], 
          "type": "Organization"
        }, 
        "familyName": "Fritz", 
        "givenName": "Mario", 
        "id": "sg:person.013361072755.17", 
        "sameAs": [
          "https://app.dimensions.ai/discover/publication?and_facet_researcher=ur.013361072755.17"
        ], 
        "type": "Person"
      }, 
      {
        "affiliation": {
          "alternateName": "Intel Visual Computing Institute, Saarbr\u00fccken, Germany", 
          "id": "http://www.grid.ac/institutes/None", 
          "name": [
            "Intel Visual Computing Institute, Saarbr\u00fccken, Germany"
          ], 
          "type": "Organization"
        }, 
        "familyName": "Grau", 
        "givenName": "Oliver", 
        "id": "sg:person.011756615143.07", 
        "sameAs": [
          "https://app.dimensions.ai/discover/publication?and_facet_researcher=ur.011756615143.07"
        ], 
        "type": "Person"
      }
    ], 
    "datePublished": "2016-08-27", 
    "datePublishedReg": "2016-08-27", 
    "description": "With the emergence of augmented reality platforms, Structure-From-Motion or visual SLAM approaches have regained in importance in order to deliver the next generation of immersive 3D experiences. As a new quality is achieved by deployment on mobile devices, computational efficiency plays an important role. In this work, we aim to reduce complexity by limiting the number of features without sacrificing quality. We select a subset of image features, using a learning based approach. A random forest is trained to pick 2D image features which are likely to be significant for a 3D reconstruction. Additionally, we aim for an objective that selects long track features, so that they can be \u201cre-used\u201d in multiple frames. We evaluate our feature selection technique on real world sequences and show a significant reduction of image features and the resulting decreased computation time is not effecting the accuracy of the 3D reconstruction.", 
    "editor": [
      {
        "familyName": "Rosenhahn", 
        "givenName": "Bodo", 
        "type": "Person"
      }, 
      {
        "familyName": "Andres", 
        "givenName": "Bjoern", 
        "type": "Person"
      }
    ], 
    "genre": "chapter", 
    "id": "sg:pub.10.1007/978-3-319-45886-1_33", 
    "isAccessibleForFree": false, 
    "isPartOf": {
      "isbn": [
        "978-3-319-45885-4", 
        "978-3-319-45886-1"
      ], 
      "name": "Pattern Recognition", 
      "type": "Book"
    }, 
    "keywords": [
      "image features", 
      "visual SLAM approach", 
      "immersive 3D experience", 
      "real-world sequences", 
      "feature selection techniques", 
      "visual SLAM", 
      "mobile devices", 
      "number of features", 
      "reality platform", 
      "SLAM approach", 
      "random forest", 
      "multiple frames", 
      "computation time", 
      "selection techniques", 
      "computational efficiency", 
      "track features", 
      "new quality", 
      "next generation", 
      "SLAM", 
      "features", 
      "deployment", 
      "platform", 
      "learning", 
      "complexity", 
      "accuracy", 
      "reconstruction", 
      "quality", 
      "frame", 
      "devices", 
      "technique", 
      "efficiency", 
      "motion", 
      "work", 
      "order", 
      "subset", 
      "generation", 
      "number", 
      "time", 
      "important role", 
      "experience", 
      "objective", 
      "sequence", 
      "forest", 
      "structure", 
      "emergence", 
      "importance", 
      "significant reduction", 
      "reduction", 
      "role", 
      "approach"
    ], 
    "name": "Learning to Select Long-Track Features for Structure-From-Motion and Visual SLAM", 
    "pagination": "402-413", 
    "productId": [
      {
        "name": "dimensions_id", 
        "type": "PropertyValue", 
        "value": [
          "pub.1019686752"
        ]
      }, 
      {
        "name": "doi", 
        "type": "PropertyValue", 
        "value": [
          "10.1007/978-3-319-45886-1_33"
        ]
      }
    ], 
    "publisher": {
      "name": "Springer Nature", 
      "type": "Organisation"
    }, 
    "sameAs": [
      "https://doi.org/10.1007/978-3-319-45886-1_33", 
      "https://app.dimensions.ai/details/publication/pub.1019686752"
    ], 
    "sdDataset": "chapters", 
    "sdDatePublished": "2022-11-24T21:14", 
    "sdLicense": "https://scigraph.springernature.com/explorer/license/", 
    "sdPublisher": {
      "name": "Springer Nature - SN SciGraph project", 
      "type": "Organization"
    }, 
    "sdSource": "s3://com-springernature-scigraph/baseset/20221124/entities/gbq_results/chapter/chapter_237.jsonl", 
    "type": "Chapter", 
    "url": "https://doi.org/10.1007/978-3-319-45886-1_33"
  }
]
 

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.10.1007/978-3-319-45886-1_33'

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

curl -H 'Accept: application/n-triples' 'https://scigraph.springernature.com/pub.10.1007/978-3-319-45886-1_33'

Turtle is a human-readable linked data format.

curl -H 'Accept: text/turtle' 'https://scigraph.springernature.com/pub.10.1007/978-3-319-45886-1_33'

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

curl -H 'Accept: application/rdf+xml' 'https://scigraph.springernature.com/pub.10.1007/978-3-319-45886-1_33'


 

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

130 TRIPLES      22 PREDICATES      74 URIs      67 LITERALS      7 BLANK NODES

Subject Predicate Object
1 sg:pub.10.1007/978-3-319-45886-1_33 schema:about anzsrc-for:08
2 anzsrc-for:0801
3 schema:author N18aa0b2f5e6c4920a6010bc005d4e6e6
4 schema:datePublished 2016-08-27
5 schema:datePublishedReg 2016-08-27
6 schema:description With the emergence of augmented reality platforms, Structure-From-Motion or visual SLAM approaches have regained in importance in order to deliver the next generation of immersive 3D experiences. As a new quality is achieved by deployment on mobile devices, computational efficiency plays an important role. In this work, we aim to reduce complexity by limiting the number of features without sacrificing quality. We select a subset of image features, using a learning based approach. A random forest is trained to pick 2D image features which are likely to be significant for a 3D reconstruction. Additionally, we aim for an objective that selects long track features, so that they can be “re-used” in multiple frames. We evaluate our feature selection technique on real world sequences and show a significant reduction of image features and the resulting decreased computation time is not effecting the accuracy of the 3D reconstruction.
7 schema:editor N617fb1d1196948348d261eb66d9df690
8 schema:genre chapter
9 schema:isAccessibleForFree false
10 schema:isPartOf N0df977a798534e959c6f68b8bd1ef49d
11 schema:keywords SLAM
12 SLAM approach
13 accuracy
14 approach
15 complexity
16 computation time
17 computational efficiency
18 deployment
19 devices
20 efficiency
21 emergence
22 experience
23 feature selection techniques
24 features
25 forest
26 frame
27 generation
28 image features
29 immersive 3D experience
30 importance
31 important role
32 learning
33 mobile devices
34 motion
35 multiple frames
36 new quality
37 next generation
38 number
39 number of features
40 objective
41 order
42 platform
43 quality
44 random forest
45 real-world sequences
46 reality platform
47 reconstruction
48 reduction
49 role
50 selection techniques
51 sequence
52 significant reduction
53 structure
54 subset
55 technique
56 time
57 track features
58 visual SLAM
59 visual SLAM approach
60 work
61 schema:name Learning to Select Long-Track Features for Structure-From-Motion and Visual SLAM
62 schema:pagination 402-413
63 schema:productId N9bd74f22310c4e3da5e3a1260d6eabc5
64 Nca44e217bbac4a909a1094eb1362ff89
65 schema:publisher N6771993aa2e34059a16477258fa66ea8
66 schema:sameAs https://app.dimensions.ai/details/publication/pub.1019686752
67 https://doi.org/10.1007/978-3-319-45886-1_33
68 schema:sdDatePublished 2022-11-24T21:14
69 schema:sdLicense https://scigraph.springernature.com/explorer/license/
70 schema:sdPublisher N6de1bf0424884ca4a8317da2835806a6
71 schema:url https://doi.org/10.1007/978-3-319-45886-1_33
72 sgo:license sg:explorer/license/
73 sgo:sdDataset chapters
74 rdf:type schema:Chapter
75 N0df977a798534e959c6f68b8bd1ef49d schema:isbn 978-3-319-45885-4
76 978-3-319-45886-1
77 schema:name Pattern Recognition
78 rdf:type schema:Book
79 N18aa0b2f5e6c4920a6010bc005d4e6e6 rdf:first N6a3cd308bb9f42d0a4a879a365931c6f
80 rdf:rest N333b2255434a4cf3aae177dd602c15da
81 N277ebed51bd34a119ca6fcae0af03123 schema:familyName Rosenhahn
82 schema:givenName Bodo
83 rdf:type schema:Person
84 N333b2255434a4cf3aae177dd602c15da rdf:first sg:person.013361072755.17
85 rdf:rest Nca1758b5d87f4c6e9a05873ddc76f35a
86 N42afc8d481934c46823f6c72b3b3ef8e rdf:first N988981282e6c4c96b3b9f442658d9cc6
87 rdf:rest rdf:nil
88 N617fb1d1196948348d261eb66d9df690 rdf:first N277ebed51bd34a119ca6fcae0af03123
89 rdf:rest N42afc8d481934c46823f6c72b3b3ef8e
90 N6771993aa2e34059a16477258fa66ea8 schema:name Springer Nature
91 rdf:type schema:Organisation
92 N6a3cd308bb9f42d0a4a879a365931c6f schema:affiliation grid-institutes:None
93 schema:familyName Scheer
94 schema:givenName Jonas
95 rdf:type schema:Person
96 N6de1bf0424884ca4a8317da2835806a6 schema:name Springer Nature - SN SciGraph project
97 rdf:type schema:Organization
98 N988981282e6c4c96b3b9f442658d9cc6 schema:familyName Andres
99 schema:givenName Bjoern
100 rdf:type schema:Person
101 N9bd74f22310c4e3da5e3a1260d6eabc5 schema:name dimensions_id
102 schema:value pub.1019686752
103 rdf:type schema:PropertyValue
104 Nca1758b5d87f4c6e9a05873ddc76f35a rdf:first sg:person.011756615143.07
105 rdf:rest rdf:nil
106 Nca44e217bbac4a909a1094eb1362ff89 schema:name doi
107 schema:value 10.1007/978-3-319-45886-1_33
108 rdf:type schema:PropertyValue
109 anzsrc-for:08 schema:inDefinedTermSet anzsrc-for:
110 schema:name Information and Computing Sciences
111 rdf:type schema:DefinedTerm
112 anzsrc-for:0801 schema:inDefinedTermSet anzsrc-for:
113 schema:name Artificial Intelligence and Image Processing
114 rdf:type schema:DefinedTerm
115 sg:person.011756615143.07 schema:affiliation grid-institutes:None
116 schema:familyName Grau
117 schema:givenName Oliver
118 schema:sameAs https://app.dimensions.ai/discover/publication?and_facet_researcher=ur.011756615143.07
119 rdf:type schema:Person
120 sg:person.013361072755.17 schema:affiliation grid-institutes:grid.419528.3
121 schema:familyName Fritz
122 schema:givenName Mario
123 schema:sameAs https://app.dimensions.ai/discover/publication?and_facet_researcher=ur.013361072755.17
124 rdf:type schema:Person
125 grid-institutes:None schema:alternateName Intel Visual Computing Institute, Saarbrücken, Germany
126 schema:name Intel Visual Computing Institute, Saarbrücken, Germany
127 rdf:type schema:Organization
128 grid-institutes:grid.419528.3 schema:alternateName Max-Planck Institute for Informatics, Saarbrücken, Germany
129 schema:name Max-Planck Institute for Informatics, Saarbrücken, Germany
130 rdf:type schema:Organization
 




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


...