A Pipeline for the Segmentation and Classification of 3D Point Clouds View Full Text


Ontology type: schema:Chapter     


Chapter Info

DATE

2014

AUTHORS

B. Douillard , J. Underwood , V. Vlaskine , A. Quadros , S. Singh

ABSTRACT

This paper presents algorithms for fast segmentation of 3D point clouds and subsequent classification of the obtained 3D segments. The method jointly determines the ground surface and segments individual objects in 3D, including overhanging structures. When compared to six other terrain modelling techniques, this approach has minimal error between the sensed data and the representation; and is fast (processing a Velodyne scan in approximately 2 seconds). Applications include improved alignment of successive scans by enabling operations in sections (Velodyne scans are aligned 7% sharper compared to an approach using raw points) and more informed decision-making (paths move around overhangs). The use of segmentation to aid classification through 3D features, such as the Spin Image or the Spherical Harmonic Descriptor, is discussed and experimentally compared. Moreover, the segmentation facilitates a novel approach to 3D classification that bypasses feature extraction and directly compares 3D shapes via the ICP algorithm. This technique is shown to achieve accuracy on par with the best feature based classifier (92.1%) while being significantly faster and allowing a clearer understanding of the classifier’s behaviour. More... »

PAGES

585-600

Identifiers

URI

http://scigraph.springernature.com/pub.10.1007/978-3-642-28572-1_40

DOI

http://dx.doi.org/10.1007/978-3-642-28572-1_40

DIMENSIONS

https://app.dimensions.ai/details/publication/pub.1053108866


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": "The Australian Centre for Field Robotics, The University of Sydney, Sydney, Australia", 
          "id": "http://www.grid.ac/institutes/grid.1013.3", 
          "name": [
            "The Australian Centre for Field Robotics, The University of Sydney, Sydney, Australia"
          ], 
          "type": "Organization"
        }, 
        "familyName": "Douillard", 
        "givenName": "B.", 
        "type": "Person"
      }, 
      {
        "affiliation": {
          "alternateName": "The Australian Centre for Field Robotics, The University of Sydney, Sydney, Australia", 
          "id": "http://www.grid.ac/institutes/grid.1013.3", 
          "name": [
            "The Australian Centre for Field Robotics, The University of Sydney, Sydney, Australia"
          ], 
          "type": "Organization"
        }, 
        "familyName": "Underwood", 
        "givenName": "J.", 
        "id": "sg:person.013552601745.79", 
        "sameAs": [
          "https://app.dimensions.ai/discover/publication?and_facet_researcher=ur.013552601745.79"
        ], 
        "type": "Person"
      }, 
      {
        "affiliation": {
          "alternateName": "The Australian Centre for Field Robotics, The University of Sydney, Sydney, Australia", 
          "id": "http://www.grid.ac/institutes/grid.1013.3", 
          "name": [
            "The Australian Centre for Field Robotics, The University of Sydney, Sydney, Australia"
          ], 
          "type": "Organization"
        }, 
        "familyName": "Vlaskine", 
        "givenName": "V.", 
        "id": "sg:person.010526133115.70", 
        "sameAs": [
          "https://app.dimensions.ai/discover/publication?and_facet_researcher=ur.010526133115.70"
        ], 
        "type": "Person"
      }, 
      {
        "affiliation": {
          "alternateName": "The Australian Centre for Field Robotics, The University of Sydney, Sydney, Australia", 
          "id": "http://www.grid.ac/institutes/grid.1013.3", 
          "name": [
            "The Australian Centre for Field Robotics, The University of Sydney, Sydney, Australia"
          ], 
          "type": "Organization"
        }, 
        "familyName": "Quadros", 
        "givenName": "A.", 
        "id": "sg:person.011362260345.69", 
        "sameAs": [
          "https://app.dimensions.ai/discover/publication?and_facet_researcher=ur.011362260345.69"
        ], 
        "type": "Person"
      }, 
      {
        "affiliation": {
          "alternateName": "The Australian Centre for Field Robotics, The University of Sydney, Sydney, Australia", 
          "id": "http://www.grid.ac/institutes/grid.1013.3", 
          "name": [
            "The Australian Centre for Field Robotics, The University of Sydney, Sydney, Australia"
          ], 
          "type": "Organization"
        }, 
        "familyName": "Singh", 
        "givenName": "S.", 
        "id": "sg:person.016506047247.09", 
        "sameAs": [
          "https://app.dimensions.ai/discover/publication?and_facet_researcher=ur.016506047247.09"
        ], 
        "type": "Person"
      }
    ], 
    "datePublished": "2014", 
    "datePublishedReg": "2014-01-01", 
    "description": "This paper presents algorithms for fast segmentation of 3D point clouds and subsequent classification of the obtained 3D segments. The method jointly determines the ground surface and segments individual objects in 3D, including overhanging structures. When compared to six other terrain modelling techniques, this approach has minimal error between the sensed data and the representation; and is fast (processing a Velodyne scan in approximately 2 seconds). Applications include improved alignment of successive scans by enabling operations in sections (Velodyne scans are aligned 7% sharper compared to an approach using raw points) and more informed decision-making (paths move around overhangs). The use of segmentation to aid classification through 3D features, such as the Spin Image or the Spherical Harmonic Descriptor, is discussed and experimentally compared. Moreover, the segmentation facilitates a novel approach to 3D classification that bypasses feature extraction and directly compares 3D shapes via the ICP algorithm. This technique is shown to achieve accuracy on par with the best feature based classifier (92.1%) while being significantly faster and allowing a clearer understanding of the classifier\u2019s behaviour.", 
    "editor": [
      {
        "familyName": "Khatib", 
        "givenName": "Oussama", 
        "type": "Person"
      }, 
      {
        "familyName": "Kumar", 
        "givenName": "Vijay", 
        "type": "Person"
      }, 
      {
        "familyName": "Sukhatme", 
        "givenName": "Gaurav", 
        "type": "Person"
      }
    ], 
    "genre": "chapter", 
    "id": "sg:pub.10.1007/978-3-642-28572-1_40", 
    "isAccessibleForFree": false, 
    "isPartOf": {
      "isbn": [
        "978-3-642-28571-4", 
        "978-3-642-28572-1"
      ], 
      "name": "Experimental Robotics", 
      "type": "Book"
    }, 
    "keywords": [
      "point clouds", 
      "use of segmentation", 
      "fast segmentation", 
      "classifier behavior", 
      "spin images", 
      "ICP algorithm", 
      "subsequent classification", 
      "segmentation", 
      "best features", 
      "individual objects", 
      "novel approach", 
      "minimal error", 
      "algorithm", 
      "cloud", 
      "classification", 
      "modelling techniques", 
      "spherical harmonic descriptors", 
      "classifier", 
      "improved alignment", 
      "ground surface", 
      "features", 
      "descriptors", 
      "images", 
      "objects", 
      "pipeline", 
      "technique", 
      "accuracy", 
      "representation", 
      "extraction", 
      "applications", 
      "error", 
      "operation", 
      "alignment", 
      "successive scans", 
      "data", 
      "par", 
      "method", 
      "behavior", 
      "surface", 
      "clear understanding", 
      "use", 
      "shape", 
      "structure", 
      "approach", 
      "segments", 
      "scans", 
      "sections", 
      "understanding", 
      "paper"
    ], 
    "name": "A Pipeline for the Segmentation and Classification of 3D Point Clouds", 
    "pagination": "585-600", 
    "productId": [
      {
        "name": "dimensions_id", 
        "type": "PropertyValue", 
        "value": [
          "pub.1053108866"
        ]
      }, 
      {
        "name": "doi", 
        "type": "PropertyValue", 
        "value": [
          "10.1007/978-3-642-28572-1_40"
        ]
      }
    ], 
    "publisher": {
      "name": "Springer Nature", 
      "type": "Organisation"
    }, 
    "sameAs": [
      "https://doi.org/10.1007/978-3-642-28572-1_40", 
      "https://app.dimensions.ai/details/publication/pub.1053108866"
    ], 
    "sdDataset": "chapters", 
    "sdDatePublished": "2022-10-01T06:57", 
    "sdLicense": "https://scigraph.springernature.com/explorer/license/", 
    "sdPublisher": {
      "name": "Springer Nature - SN SciGraph project", 
      "type": "Organization"
    }, 
    "sdSource": "s3://com-springernature-scigraph/baseset/20221001/entities/gbq_results/chapter/chapter_332.jsonl", 
    "type": "Chapter", 
    "url": "https://doi.org/10.1007/978-3-642-28572-1_40"
  }
]
 

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.10.1007/978-3-642-28572-1_40'

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

curl -H 'Accept: application/n-triples' 'https://scigraph.springernature.com/pub.10.1007/978-3-642-28572-1_40'

Turtle is a human-readable linked data format.

curl -H 'Accept: text/turtle' 'https://scigraph.springernature.com/pub.10.1007/978-3-642-28572-1_40'

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

curl -H 'Accept: application/rdf+xml' 'https://scigraph.springernature.com/pub.10.1007/978-3-642-28572-1_40'


 

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

145 TRIPLES      22 PREDICATES      74 URIs      67 LITERALS      7 BLANK NODES

Subject Predicate Object
1 sg:pub.10.1007/978-3-642-28572-1_40 schema:about anzsrc-for:08
2 anzsrc-for:0801
3 schema:author N115f004e5f244fd691d4a1b4609f1331
4 schema:datePublished 2014
5 schema:datePublishedReg 2014-01-01
6 schema:description This paper presents algorithms for fast segmentation of 3D point clouds and subsequent classification of the obtained 3D segments. The method jointly determines the ground surface and segments individual objects in 3D, including overhanging structures. When compared to six other terrain modelling techniques, this approach has minimal error between the sensed data and the representation; and is fast (processing a Velodyne scan in approximately 2 seconds). Applications include improved alignment of successive scans by enabling operations in sections (Velodyne scans are aligned 7% sharper compared to an approach using raw points) and more informed decision-making (paths move around overhangs). The use of segmentation to aid classification through 3D features, such as the Spin Image or the Spherical Harmonic Descriptor, is discussed and experimentally compared. Moreover, the segmentation facilitates a novel approach to 3D classification that bypasses feature extraction and directly compares 3D shapes via the ICP algorithm. This technique is shown to achieve accuracy on par with the best feature based classifier (92.1%) while being significantly faster and allowing a clearer understanding of the classifier’s behaviour.
7 schema:editor N53bed3c2c30345a0920d169b7fee2b8c
8 schema:genre chapter
9 schema:isAccessibleForFree false
10 schema:isPartOf Nf7fe739d02cd45fbbb1042a338b46cdb
11 schema:keywords ICP algorithm
12 accuracy
13 algorithm
14 alignment
15 applications
16 approach
17 behavior
18 best features
19 classification
20 classifier
21 classifier behavior
22 clear understanding
23 cloud
24 data
25 descriptors
26 error
27 extraction
28 fast segmentation
29 features
30 ground surface
31 images
32 improved alignment
33 individual objects
34 method
35 minimal error
36 modelling techniques
37 novel approach
38 objects
39 operation
40 paper
41 par
42 pipeline
43 point clouds
44 representation
45 scans
46 sections
47 segmentation
48 segments
49 shape
50 spherical harmonic descriptors
51 spin images
52 structure
53 subsequent classification
54 successive scans
55 surface
56 technique
57 understanding
58 use
59 use of segmentation
60 schema:name A Pipeline for the Segmentation and Classification of 3D Point Clouds
61 schema:pagination 585-600
62 schema:productId N55100b745b684f21b91734d0d08bc11b
63 N70b4682f36b343acb5dce6437baf96c2
64 schema:publisher Ncfd86ec5d1da400cb1f3ae4734a6f78e
65 schema:sameAs https://app.dimensions.ai/details/publication/pub.1053108866
66 https://doi.org/10.1007/978-3-642-28572-1_40
67 schema:sdDatePublished 2022-10-01T06:57
68 schema:sdLicense https://scigraph.springernature.com/explorer/license/
69 schema:sdPublisher N00500d12993f40b29bb9cb64aa827c44
70 schema:url https://doi.org/10.1007/978-3-642-28572-1_40
71 sgo:license sg:explorer/license/
72 sgo:sdDataset chapters
73 rdf:type schema:Chapter
74 N00500d12993f40b29bb9cb64aa827c44 schema:name Springer Nature - SN SciGraph project
75 rdf:type schema:Organization
76 N02054f70ca0a45b4939dc75f67417690 rdf:first sg:person.010526133115.70
77 rdf:rest N329c8e2574cd4796a42d95829f5f394f
78 N115f004e5f244fd691d4a1b4609f1331 rdf:first N6656bd20b22e49dba75679ba77119913
79 rdf:rest N2c14300a5efe4ea781d7ee30aab9357e
80 N22c1888f28df45abbe0ab2d9356aae7c schema:familyName Khatib
81 schema:givenName Oussama
82 rdf:type schema:Person
83 N2c14300a5efe4ea781d7ee30aab9357e rdf:first sg:person.013552601745.79
84 rdf:rest N02054f70ca0a45b4939dc75f67417690
85 N329c8e2574cd4796a42d95829f5f394f rdf:first sg:person.011362260345.69
86 rdf:rest Ne9f5a2b6870f4c4883ab4ea0922f3094
87 N3cb3f5ee568f49aca1872e9b0e2974ec rdf:first N9a83a7d13b5b40c68c12794ff39ec4b7
88 rdf:rest rdf:nil
89 N43fa171a14504903afd7dfd1ca2a222e rdf:first Na5cc8af65f1b42828183ead2655a8578
90 rdf:rest N3cb3f5ee568f49aca1872e9b0e2974ec
91 N53bed3c2c30345a0920d169b7fee2b8c rdf:first N22c1888f28df45abbe0ab2d9356aae7c
92 rdf:rest N43fa171a14504903afd7dfd1ca2a222e
93 N55100b745b684f21b91734d0d08bc11b schema:name dimensions_id
94 schema:value pub.1053108866
95 rdf:type schema:PropertyValue
96 N6656bd20b22e49dba75679ba77119913 schema:affiliation grid-institutes:grid.1013.3
97 schema:familyName Douillard
98 schema:givenName B.
99 rdf:type schema:Person
100 N70b4682f36b343acb5dce6437baf96c2 schema:name doi
101 schema:value 10.1007/978-3-642-28572-1_40
102 rdf:type schema:PropertyValue
103 N9a83a7d13b5b40c68c12794ff39ec4b7 schema:familyName Sukhatme
104 schema:givenName Gaurav
105 rdf:type schema:Person
106 Na5cc8af65f1b42828183ead2655a8578 schema:familyName Kumar
107 schema:givenName Vijay
108 rdf:type schema:Person
109 Ncfd86ec5d1da400cb1f3ae4734a6f78e schema:name Springer Nature
110 rdf:type schema:Organisation
111 Ne9f5a2b6870f4c4883ab4ea0922f3094 rdf:first sg:person.016506047247.09
112 rdf:rest rdf:nil
113 Nf7fe739d02cd45fbbb1042a338b46cdb schema:isbn 978-3-642-28571-4
114 978-3-642-28572-1
115 schema:name Experimental Robotics
116 rdf:type schema:Book
117 anzsrc-for:08 schema:inDefinedTermSet anzsrc-for:
118 schema:name Information and Computing Sciences
119 rdf:type schema:DefinedTerm
120 anzsrc-for:0801 schema:inDefinedTermSet anzsrc-for:
121 schema:name Artificial Intelligence and Image Processing
122 rdf:type schema:DefinedTerm
123 sg:person.010526133115.70 schema:affiliation grid-institutes:grid.1013.3
124 schema:familyName Vlaskine
125 schema:givenName V.
126 schema:sameAs https://app.dimensions.ai/discover/publication?and_facet_researcher=ur.010526133115.70
127 rdf:type schema:Person
128 sg:person.011362260345.69 schema:affiliation grid-institutes:grid.1013.3
129 schema:familyName Quadros
130 schema:givenName A.
131 schema:sameAs https://app.dimensions.ai/discover/publication?and_facet_researcher=ur.011362260345.69
132 rdf:type schema:Person
133 sg:person.013552601745.79 schema:affiliation grid-institutes:grid.1013.3
134 schema:familyName Underwood
135 schema:givenName J.
136 schema:sameAs https://app.dimensions.ai/discover/publication?and_facet_researcher=ur.013552601745.79
137 rdf:type schema:Person
138 sg:person.016506047247.09 schema:affiliation grid-institutes:grid.1013.3
139 schema:familyName Singh
140 schema:givenName S.
141 schema:sameAs https://app.dimensions.ai/discover/publication?and_facet_researcher=ur.016506047247.09
142 rdf:type schema:Person
143 grid-institutes:grid.1013.3 schema:alternateName The Australian Centre for Field Robotics, The University of Sydney, Sydney, Australia
144 schema:name The Australian Centre for Field Robotics, The University of Sydney, Sydney, Australia
145 rdf:type schema:Organization
 




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


...