A novel point cloud registration using 2D image features View Full Text


Ontology type: schema:ScholarlyArticle      Open Access: True


Article Info

DATE

2017-01-07

AUTHORS

Chien-Chou Lin, Yen-Chou Tai, Jhong-Jin Lee, Yong-Sheng Chen

ABSTRACT

Since a 3D scanner only captures a scene of a 3D object at a time, a 3D registration for multi-scene is the key issue of 3D modeling. This paper presents a novel and an efficient 3D registration method based on 2D local feature matching. The proposed method transforms the point clouds into 2D bearing angle images and then uses the 2D feature based matching method, SURF, to find matching pixel pairs between two images. The corresponding points of 3D point clouds can be obtained by those pixel pairs. Since the corresponding pairs are sorted by their distance between matching features, only the top half of the corresponding pairs are used to find the optimal rotation matrix by the least squares approximation. In this paper, the optimal rotation matrix is derived by orthogonal Procrustes method (SVD-based approach). Therefore, the 3D model of an object can be reconstructed by aligning those point clouds with the optimal transformation matrix. Experimental results show that the accuracy of the proposed method is close to the ICP, but the computation cost is reduced significantly. The performance is six times faster than the generalized-ICP algorithm. Furthermore, while the ICP requires high alignment similarity of two scenes, the proposed method is robust to a larger difference of viewing angle. More... »

PAGES

5

References to SciGraph publications

  • 1994-10. Iterative point matching for registration of free-form curves and surfaces in INTERNATIONAL JOURNAL OF COMPUTER VISION
  • 2006. SURF: Speeded Up Robust Features in COMPUTER VISION – ECCV 2006
  • Identifiers

    URI

    http://scigraph.springernature.com/pub.10.1186/s13634-016-0435-y

    DOI

    http://dx.doi.org/10.1186/s13634-016-0435-y

    DIMENSIONS

    https://app.dimensions.ai/details/publication/pub.1018039364


    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 and Information Engineering, National Yunlin University of Science and Technology, Yunlin, Taiwan, ROC", 
              "id": "http://www.grid.ac/institutes/grid.412127.3", 
              "name": [
                "Department of Computer Science and Information Engineering, National Yunlin University of Science and Technology, Yunlin, Taiwan, ROC"
              ], 
              "type": "Organization"
            }, 
            "familyName": "Lin", 
            "givenName": "Chien-Chou", 
            "id": "sg:person.013513003635.72", 
            "sameAs": [
              "https://app.dimensions.ai/discover/publication?and_facet_researcher=ur.013513003635.72"
            ], 
            "type": "Person"
          }, 
          {
            "affiliation": {
              "alternateName": "Department of Computer Science, National Chiao Tung University, Hsinchou, Taiwan, ROC", 
              "id": "http://www.grid.ac/institutes/grid.260539.b", 
              "name": [
                "Department of Computer Science, National Chiao Tung University, Hsinchou, Taiwan, ROC"
              ], 
              "type": "Organization"
            }, 
            "familyName": "Tai", 
            "givenName": "Yen-Chou", 
            "id": "sg:person.016234277457.41", 
            "sameAs": [
              "https://app.dimensions.ai/discover/publication?and_facet_researcher=ur.016234277457.41"
            ], 
            "type": "Person"
          }, 
          {
            "affiliation": {
              "alternateName": "Department of Computer Science and Information Engineering, National Yunlin University of Science and Technology, Yunlin, Taiwan, ROC", 
              "id": "http://www.grid.ac/institutes/grid.412127.3", 
              "name": [
                "Department of Computer Science and Information Engineering, National Yunlin University of Science and Technology, Yunlin, Taiwan, ROC"
              ], 
              "type": "Organization"
            }, 
            "familyName": "Lee", 
            "givenName": "Jhong-Jin", 
            "type": "Person"
          }, 
          {
            "affiliation": {
              "alternateName": "Department of Computer Science, National Chiao Tung University, Hsinchou, Taiwan, ROC", 
              "id": "http://www.grid.ac/institutes/grid.260539.b", 
              "name": [
                "Department of Computer Science, National Chiao Tung University, Hsinchou, Taiwan, ROC"
              ], 
              "type": "Organization"
            }, 
            "familyName": "Chen", 
            "givenName": "Yong-Sheng", 
            "id": "sg:person.01336567711.81", 
            "sameAs": [
              "https://app.dimensions.ai/discover/publication?and_facet_researcher=ur.01336567711.81"
            ], 
            "type": "Person"
          }
        ], 
        "citation": [
          {
            "id": "sg:pub.10.1007/bf01427149", 
            "sameAs": [
              "https://app.dimensions.ai/details/publication/pub.1050436308", 
              "https://doi.org/10.1007/bf01427149"
            ], 
            "type": "CreativeWork"
          }, 
          {
            "id": "sg:pub.10.1007/11744023_32", 
            "sameAs": [
              "https://app.dimensions.ai/details/publication/pub.1000158525", 
              "https://doi.org/10.1007/11744023_32"
            ], 
            "type": "CreativeWork"
          }
        ], 
        "datePublished": "2017-01-07", 
        "datePublishedReg": "2017-01-07", 
        "description": "Since a 3D scanner only captures a scene of a 3D object at a time, a 3D registration for multi-scene is the key issue of 3D modeling. This paper presents a novel and an efficient 3D registration method based on 2D local feature matching. The proposed method transforms the point clouds into 2D bearing angle images and then uses the 2D feature based matching method, SURF, to find matching pixel pairs between two images. The corresponding points of 3D point clouds can be obtained by those pixel pairs. Since the corresponding pairs are sorted by their distance between matching features, only the top half of the corresponding pairs are used to find the optimal rotation matrix by the least squares approximation. In this paper, the optimal rotation matrix is derived by orthogonal Procrustes method (SVD-based approach). Therefore, the 3D model of an object can be reconstructed by aligning those point clouds with the optimal transformation matrix. Experimental results show that the accuracy of the proposed method is close to the ICP, but the computation cost is reduced significantly. The performance is six times faster than the generalized-ICP algorithm. Furthermore, while the ICP requires high alignment similarity of two scenes, the proposed method is robust to a larger difference of viewing angle.", 
        "genre": "article", 
        "id": "sg:pub.10.1186/s13634-016-0435-y", 
        "isAccessibleForFree": true, 
        "isPartOf": [
          {
            "id": "sg:journal.1371279", 
            "issn": [
              "1687-6172", 
              "1687-0433"
            ], 
            "name": "EURASIP Journal on Advances in Signal Processing", 
            "publisher": "Springer Nature", 
            "type": "Periodical"
          }, 
          {
            "issueNumber": "1", 
            "type": "PublicationIssue"
          }, 
          {
            "type": "PublicationVolume", 
            "volumeNumber": "2017"
          }
        ], 
        "keywords": [
          "point clouds", 
          "pixel pairs", 
          "local feature matching", 
          "point cloud registration", 
          "feature matching", 
          "image features", 
          "computation cost", 
          "matching features", 
          "optimal transformation matrix", 
          "cloud registration", 
          "alignment similarity", 
          "registration method", 
          "matching method", 
          "corresponding points", 
          "optimal rotation matrix", 
          "rotation matrix", 
          "cloud", 
          "angle images", 
          "orthogonal Procrustes method", 
          "experimental results", 
          "scene", 
          "key issues", 
          "corresponding pair", 
          "transformation matrix", 
          "images", 
          "objects", 
          "registration", 
          "SURF", 
          "algorithm", 
          "features", 
          "matching", 
          "method", 
          "accuracy", 
          "least squares approximation", 
          "performance", 
          "cost", 
          "modeling", 
          "scanner", 
          "issues", 
          "squares approximation", 
          "Procrustes method", 
          "time", 
          "pairs", 
          "model", 
          "similarity", 
          "matrix", 
          "point", 
          "distance", 
          "novel", 
          "results", 
          "ICP", 
          "approximation", 
          "angle", 
          "top half", 
          "large differences", 
          "differences", 
          "half", 
          "paper"
        ], 
        "name": "A novel point cloud registration using 2D image features", 
        "pagination": "5", 
        "productId": [
          {
            "name": "dimensions_id", 
            "type": "PropertyValue", 
            "value": [
              "pub.1018039364"
            ]
          }, 
          {
            "name": "doi", 
            "type": "PropertyValue", 
            "value": [
              "10.1186/s13634-016-0435-y"
            ]
          }
        ], 
        "sameAs": [
          "https://doi.org/10.1186/s13634-016-0435-y", 
          "https://app.dimensions.ai/details/publication/pub.1018039364"
        ], 
        "sdDataset": "articles", 
        "sdDatePublished": "2022-08-04T17:04", 
        "sdLicense": "https://scigraph.springernature.com/explorer/license/", 
        "sdPublisher": {
          "name": "Springer Nature - SN SciGraph project", 
          "type": "Organization"
        }, 
        "sdSource": "s3://com-springernature-scigraph/baseset/20220804/entities/gbq_results/article/article_724.jsonl", 
        "type": "ScholarlyArticle", 
        "url": "https://doi.org/10.1186/s13634-016-0435-y"
      }
    ]
     

    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.10.1186/s13634-016-0435-y'

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

    curl -H 'Accept: application/n-triples' 'https://scigraph.springernature.com/pub.10.1186/s13634-016-0435-y'

    Turtle is a human-readable linked data format.

    curl -H 'Accept: text/turtle' 'https://scigraph.springernature.com/pub.10.1186/s13634-016-0435-y'

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

    curl -H 'Accept: application/rdf+xml' 'https://scigraph.springernature.com/pub.10.1186/s13634-016-0435-y'


     

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

    146 TRIPLES      21 PREDICATES      84 URIs      74 LITERALS      6 BLANK NODES

    Subject Predicate Object
    1 sg:pub.10.1186/s13634-016-0435-y schema:about anzsrc-for:08
    2 anzsrc-for:0801
    3 schema:author N9754e6af47c3435d818d02c46e407895
    4 schema:citation sg:pub.10.1007/11744023_32
    5 sg:pub.10.1007/bf01427149
    6 schema:datePublished 2017-01-07
    7 schema:datePublishedReg 2017-01-07
    8 schema:description Since a 3D scanner only captures a scene of a 3D object at a time, a 3D registration for multi-scene is the key issue of 3D modeling. This paper presents a novel and an efficient 3D registration method based on 2D local feature matching. The proposed method transforms the point clouds into 2D bearing angle images and then uses the 2D feature based matching method, SURF, to find matching pixel pairs between two images. The corresponding points of 3D point clouds can be obtained by those pixel pairs. Since the corresponding pairs are sorted by their distance between matching features, only the top half of the corresponding pairs are used to find the optimal rotation matrix by the least squares approximation. In this paper, the optimal rotation matrix is derived by orthogonal Procrustes method (SVD-based approach). Therefore, the 3D model of an object can be reconstructed by aligning those point clouds with the optimal transformation matrix. Experimental results show that the accuracy of the proposed method is close to the ICP, but the computation cost is reduced significantly. The performance is six times faster than the generalized-ICP algorithm. Furthermore, while the ICP requires high alignment similarity of two scenes, the proposed method is robust to a larger difference of viewing angle.
    9 schema:genre article
    10 schema:isAccessibleForFree true
    11 schema:isPartOf Nb0b3ff07c7774be8b4c4420abc3ba735
    12 Nfafe8f92b7c240878fcdf2cc05ef3bdd
    13 sg:journal.1371279
    14 schema:keywords ICP
    15 Procrustes method
    16 SURF
    17 accuracy
    18 algorithm
    19 alignment similarity
    20 angle
    21 angle images
    22 approximation
    23 cloud
    24 cloud registration
    25 computation cost
    26 corresponding pair
    27 corresponding points
    28 cost
    29 differences
    30 distance
    31 experimental results
    32 feature matching
    33 features
    34 half
    35 image features
    36 images
    37 issues
    38 key issues
    39 large differences
    40 least squares approximation
    41 local feature matching
    42 matching
    43 matching features
    44 matching method
    45 matrix
    46 method
    47 model
    48 modeling
    49 novel
    50 objects
    51 optimal rotation matrix
    52 optimal transformation matrix
    53 orthogonal Procrustes method
    54 pairs
    55 paper
    56 performance
    57 pixel pairs
    58 point
    59 point cloud registration
    60 point clouds
    61 registration
    62 registration method
    63 results
    64 rotation matrix
    65 scanner
    66 scene
    67 similarity
    68 squares approximation
    69 time
    70 top half
    71 transformation matrix
    72 schema:name A novel point cloud registration using 2D image features
    73 schema:pagination 5
    74 schema:productId N742bf26710874614b829b93e842f0e50
    75 N88ce38fbb39c46bbb64579b8fe97c334
    76 schema:sameAs https://app.dimensions.ai/details/publication/pub.1018039364
    77 https://doi.org/10.1186/s13634-016-0435-y
    78 schema:sdDatePublished 2022-08-04T17:04
    79 schema:sdLicense https://scigraph.springernature.com/explorer/license/
    80 schema:sdPublisher N4495676d29494908bf3043087e5e3b62
    81 schema:url https://doi.org/10.1186/s13634-016-0435-y
    82 sgo:license sg:explorer/license/
    83 sgo:sdDataset articles
    84 rdf:type schema:ScholarlyArticle
    85 N20953d137ddc4cedbb65bd202d54144b schema:affiliation grid-institutes:grid.412127.3
    86 schema:familyName Lee
    87 schema:givenName Jhong-Jin
    88 rdf:type schema:Person
    89 N23720c93328a40f7a6660c2050539eca rdf:first sg:person.016234277457.41
    90 rdf:rest N892cb35b6ad0489fb5ec2c9c1bb494b2
    91 N4495676d29494908bf3043087e5e3b62 schema:name Springer Nature - SN SciGraph project
    92 rdf:type schema:Organization
    93 N56814d5a9df04fbda24afd8b457e233d rdf:first sg:person.01336567711.81
    94 rdf:rest rdf:nil
    95 N742bf26710874614b829b93e842f0e50 schema:name doi
    96 schema:value 10.1186/s13634-016-0435-y
    97 rdf:type schema:PropertyValue
    98 N88ce38fbb39c46bbb64579b8fe97c334 schema:name dimensions_id
    99 schema:value pub.1018039364
    100 rdf:type schema:PropertyValue
    101 N892cb35b6ad0489fb5ec2c9c1bb494b2 rdf:first N20953d137ddc4cedbb65bd202d54144b
    102 rdf:rest N56814d5a9df04fbda24afd8b457e233d
    103 N9754e6af47c3435d818d02c46e407895 rdf:first sg:person.013513003635.72
    104 rdf:rest N23720c93328a40f7a6660c2050539eca
    105 Nb0b3ff07c7774be8b4c4420abc3ba735 schema:issueNumber 1
    106 rdf:type schema:PublicationIssue
    107 Nfafe8f92b7c240878fcdf2cc05ef3bdd schema:volumeNumber 2017
    108 rdf:type schema:PublicationVolume
    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:journal.1371279 schema:issn 1687-0433
    116 1687-6172
    117 schema:name EURASIP Journal on Advances in Signal Processing
    118 schema:publisher Springer Nature
    119 rdf:type schema:Periodical
    120 sg:person.01336567711.81 schema:affiliation grid-institutes:grid.260539.b
    121 schema:familyName Chen
    122 schema:givenName Yong-Sheng
    123 schema:sameAs https://app.dimensions.ai/discover/publication?and_facet_researcher=ur.01336567711.81
    124 rdf:type schema:Person
    125 sg:person.013513003635.72 schema:affiliation grid-institutes:grid.412127.3
    126 schema:familyName Lin
    127 schema:givenName Chien-Chou
    128 schema:sameAs https://app.dimensions.ai/discover/publication?and_facet_researcher=ur.013513003635.72
    129 rdf:type schema:Person
    130 sg:person.016234277457.41 schema:affiliation grid-institutes:grid.260539.b
    131 schema:familyName Tai
    132 schema:givenName Yen-Chou
    133 schema:sameAs https://app.dimensions.ai/discover/publication?and_facet_researcher=ur.016234277457.41
    134 rdf:type schema:Person
    135 sg:pub.10.1007/11744023_32 schema:sameAs https://app.dimensions.ai/details/publication/pub.1000158525
    136 https://doi.org/10.1007/11744023_32
    137 rdf:type schema:CreativeWork
    138 sg:pub.10.1007/bf01427149 schema:sameAs https://app.dimensions.ai/details/publication/pub.1050436308
    139 https://doi.org/10.1007/bf01427149
    140 rdf:type schema:CreativeWork
    141 grid-institutes:grid.260539.b schema:alternateName Department of Computer Science, National Chiao Tung University, Hsinchou, Taiwan, ROC
    142 schema:name Department of Computer Science, National Chiao Tung University, Hsinchou, Taiwan, ROC
    143 rdf:type schema:Organization
    144 grid-institutes:grid.412127.3 schema:alternateName Department of Computer Science and Information Engineering, National Yunlin University of Science and Technology, Yunlin, Taiwan, ROC
    145 schema:name Department of Computer Science and Information Engineering, National Yunlin University of Science and Technology, Yunlin, Taiwan, ROC
    146 rdf:type schema:Organization
     




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


    ...