3-D Vision and Recognition View Full Text


Ontology type: schema:Chapter     


Chapter Info

DATE

2008

AUTHORS

Kostas Daniilidis , Jan-Olof Eklundh

ABSTRACT

In this chapter, we describe methods to be applied on a robot equipped with one or more camera sensors. Our goal is to present representations and models for both three-dimensional (3-D) motion and structure estimation as well as recognition. We do not delve into estimation and inference issues since these are extensively treated in other chapters. The same applies to the fusion with other sensors, which we heavily encourage but do not describe here. In the first part we describe the main methods in 3-D inference from two-dimensional (2-D) images. We are at the point where we could propose a recipe, at least for a small spatial extent. If we are able to track a few visual features in our images, we are able to estimate the self-motion of the robot as well as its pose with respect to any known landmark. Having solutions for minimal case problems, the obvious way here is to apply random sample consensus. If no known 3-D landmark is given then the trajectory of the camera exhibits drift. From the trajectory of the camera, time windows over several frames are selected and a 3-D dense depth map is obtained through solving the stereo problem. Large-scale reconstructions based on camera only do raise challenges with respect to drift and loop closing. In the second part we deal with recognition as appealed to robotics. The main challenge here is to detect an instance of an object and recognize or categorize it. Since in robotics applications an object of interest always resides in a cluttered environment any algorithm has to be insensitive to missing parts of the object of interest and outliers. The dominant paradigm is based on matching the appearance of pictures. Features are detected and quantized into visual words. Similarity is based on the difference between histograms of such visual words. Recognition has a long way to go but robotics provides the opportunity to explore an object and be active in the recognition process. More... »

PAGES

543-562

Identifiers

URI

http://scigraph.springernature.com/pub.10.1007/978-3-540-30301-5_24

DOI

http://dx.doi.org/10.1007/978-3-540-30301-5_24

DIMENSIONS

https://app.dimensions.ai/details/publication/pub.1029772224


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 and Information Science GRASP Laboratory, University of Pennsylvania, 3330 Walnut Street, 19104, Philadelphia, PA, USA", 
          "id": "http://www.grid.ac/institutes/grid.25879.31", 
          "name": [
            "Department of Computer and Information Science GRASP Laboratory, University of Pennsylvania, 3330 Walnut Street, 19104, Philadelphia, PA, USA"
          ], 
          "type": "Organization"
        }, 
        "familyName": "Daniilidis", 
        "givenName": "Kostas", 
        "id": "sg:person.0643156134.22", 
        "sameAs": [
          "https://app.dimensions.ai/discover/publication?and_facet_researcher=ur.0643156134.22"
        ], 
        "type": "Person"
      }, 
      {
        "affiliation": {
          "alternateName": "KTH Royal Institute of Technology, Teknikringen 14, 10044, Stockholm, Sweden", 
          "id": "http://www.grid.ac/institutes/grid.5037.1", 
          "name": [
            "KTH Royal Institute of Technology, Teknikringen 14, 10044, Stockholm, Sweden"
          ], 
          "type": "Organization"
        }, 
        "familyName": "Eklundh", 
        "givenName": "Jan-Olof", 
        "id": "sg:person.014400652155.17", 
        "sameAs": [
          "https://app.dimensions.ai/discover/publication?and_facet_researcher=ur.014400652155.17"
        ], 
        "type": "Person"
      }
    ], 
    "datePublished": "2008", 
    "datePublishedReg": "2008-01-01", 
    "description": "Abstract\n              In this chapter, we describe methods to be applied on a\u00a0robot equipped with one or more camera sensors. Our goal is to present representations and models for both three-dimensional (3-D) motion and structure estimation as well as recognition. We do not delve into estimation and inference issues since these are extensively treated in other chapters. The same applies to the fusion with other sensors, which we heavily encourage but do not describe here.\n            \n              In the first part we describe the main methods in 3-D inference from two-dimensional (2-D) images. We are at the point where we could propose a\u00a0recipe, at least for a\u00a0small spatial extent. If we are able to track a\u00a0few visual features in our images, we are able to estimate the self-motion of the robot as well as its pose with respect to any known landmark. Having solutions for minimal case problems, the obvious way here is to apply random sample consensus. If no known 3-D landmark is given then the trajectory of the camera exhibits drift. From the trajectory of the camera, time windows over several frames are selected and a\u00a03-D dense depth map is obtained through solving the stereo problem. Large-scale reconstructions based on camera only do raise challenges with respect to drift and loop closing.\n            \n              In the second part we deal with recognition as appealed to robotics. The main challenge here is to detect an instance of an object and recognize or categorize it. Since in robotics applications an object of interest always resides in a\u00a0cluttered environment any algorithm has to be insensitive to missing parts of the object of interest and outliers. The dominant paradigm is based on matching the appearance of pictures. Features are detected and quantized into visual words. Similarity is based on the difference between histograms of such visual words. Recognition has a\u00a0long way to go but robotics provides the opportunity to explore an object and be active in the recognition process.", 
    "editor": [
      {
        "familyName": "Siciliano", 
        "givenName": "Bruno", 
        "type": "Person"
      }, 
      {
        "familyName": "Khatib", 
        "givenName": "Oussama", 
        "type": "Person"
      }
    ], 
    "genre": "chapter", 
    "id": "sg:pub.10.1007/978-3-540-30301-5_24", 
    "isAccessibleForFree": false, 
    "isPartOf": {
      "isbn": [
        "978-3-540-23957-4", 
        "978-3-540-30301-5"
      ], 
      "name": "Springer Handbook of Robotics", 
      "type": "Book"
    }, 
    "keywords": [
      "object of interest", 
      "visual words", 
      "dense depth map", 
      "random sample consensus", 
      "camera sensor", 
      "depth map", 
      "robotic applications", 
      "visual features", 
      "stereo problem", 
      "three-dimensional motion", 
      "loop closing", 
      "sample consensus", 
      "two-dimensional images", 
      "large-scale reconstruction", 
      "inference issues", 
      "recognition process", 
      "structure estimation", 
      "camera", 
      "robot", 
      "robotics", 
      "main challenges", 
      "case problem", 
      "objects", 
      "recognition", 
      "images", 
      "time window", 
      "estimation", 
      "pose", 
      "sensors", 
      "trajectories", 
      "landmarks", 
      "algorithm", 
      "problem", 
      "obvious way", 
      "second part", 
      "histogram", 
      "challenges", 
      "first part", 
      "main methods", 
      "features", 
      "vision", 
      "dominant paradigm", 
      "motion", 
      "small spatial extent", 
      "outliers", 
      "paradigm", 
      "long way", 
      "inference", 
      "representation", 
      "words", 
      "way", 
      "spatial extent", 
      "instances", 
      "environment", 
      "solution", 
      "fusion", 
      "frame", 
      "applications", 
      "method", 
      "maps", 
      "issues", 
      "respect", 
      "goal", 
      "interest", 
      "model", 
      "drift", 
      "reconstruction", 
      "part", 
      "point", 
      "chapter", 
      "window", 
      "similarity", 
      "opportunities", 
      "process", 
      "recipes", 
      "picture", 
      "consensus", 
      "closing", 
      "appearance", 
      "extent", 
      "differences"
    ], 
    "name": "3-D Vision and Recognition", 
    "pagination": "543-562", 
    "productId": [
      {
        "name": "dimensions_id", 
        "type": "PropertyValue", 
        "value": [
          "pub.1029772224"
        ]
      }, 
      {
        "name": "doi", 
        "type": "PropertyValue", 
        "value": [
          "10.1007/978-3-540-30301-5_24"
        ]
      }
    ], 
    "publisher": {
      "name": "Springer Nature", 
      "type": "Organisation"
    }, 
    "sameAs": [
      "https://doi.org/10.1007/978-3-540-30301-5_24", 
      "https://app.dimensions.ai/details/publication/pub.1029772224"
    ], 
    "sdDataset": "chapters", 
    "sdDatePublished": "2022-12-01T06:55", 
    "sdLicense": "https://scigraph.springernature.com/explorer/license/", 
    "sdPublisher": {
      "name": "Springer Nature - SN SciGraph project", 
      "type": "Organization"
    }, 
    "sdSource": "s3://com-springernature-scigraph/baseset/20221201/entities/gbq_results/chapter/chapter_92.jsonl", 
    "type": "Chapter", 
    "url": "https://doi.org/10.1007/978-3-540-30301-5_24"
  }
]
 

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.10.1007/978-3-540-30301-5_24'

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

curl -H 'Accept: application/n-triples' 'https://scigraph.springernature.com/pub.10.1007/978-3-540-30301-5_24'

Turtle is a human-readable linked data format.

curl -H 'Accept: text/turtle' 'https://scigraph.springernature.com/pub.10.1007/978-3-540-30301-5_24'

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

curl -H 'Accept: application/rdf+xml' 'https://scigraph.springernature.com/pub.10.1007/978-3-540-30301-5_24'


 

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

155 TRIPLES      22 PREDICATES      105 URIs      98 LITERALS      7 BLANK NODES

Subject Predicate Object
1 sg:pub.10.1007/978-3-540-30301-5_24 schema:about anzsrc-for:08
2 anzsrc-for:0801
3 schema:author N328f078036634d6c8efcb394163bb4dc
4 schema:datePublished 2008
5 schema:datePublishedReg 2008-01-01
6 schema:description Abstract In this chapter, we describe methods to be applied on a robot equipped with one or more camera sensors. Our goal is to present representations and models for both three-dimensional (3-D) motion and structure estimation as well as recognition. We do not delve into estimation and inference issues since these are extensively treated in other chapters. The same applies to the fusion with other sensors, which we heavily encourage but do not describe here. In the first part we describe the main methods in 3-D inference from two-dimensional (2-D) images. We are at the point where we could propose a recipe, at least for a small spatial extent. If we are able to track a few visual features in our images, we are able to estimate the self-motion of the robot as well as its pose with respect to any known landmark. Having solutions for minimal case problems, the obvious way here is to apply random sample consensus. If no known 3-D landmark is given then the trajectory of the camera exhibits drift. From the trajectory of the camera, time windows over several frames are selected and a 3-D dense depth map is obtained through solving the stereo problem. Large-scale reconstructions based on camera only do raise challenges with respect to drift and loop closing. In the second part we deal with recognition as appealed to robotics. The main challenge here is to detect an instance of an object and recognize or categorize it. Since in robotics applications an object of interest always resides in a cluttered environment any algorithm has to be insensitive to missing parts of the object of interest and outliers. The dominant paradigm is based on matching the appearance of pictures. Features are detected and quantized into visual words. Similarity is based on the difference between histograms of such visual words. Recognition has a long way to go but robotics provides the opportunity to explore an object and be active in the recognition process.
7 schema:editor N858022c858244f6f81ef24ff40ac6daa
8 schema:genre chapter
9 schema:isAccessibleForFree false
10 schema:isPartOf N1b6cc67d2d174bc78e678b35cb9f41f7
11 schema:keywords algorithm
12 appearance
13 applications
14 camera
15 camera sensor
16 case problem
17 challenges
18 chapter
19 closing
20 consensus
21 dense depth map
22 depth map
23 differences
24 dominant paradigm
25 drift
26 environment
27 estimation
28 extent
29 features
30 first part
31 frame
32 fusion
33 goal
34 histogram
35 images
36 inference
37 inference issues
38 instances
39 interest
40 issues
41 landmarks
42 large-scale reconstruction
43 long way
44 loop closing
45 main challenges
46 main methods
47 maps
48 method
49 model
50 motion
51 object of interest
52 objects
53 obvious way
54 opportunities
55 outliers
56 paradigm
57 part
58 picture
59 point
60 pose
61 problem
62 process
63 random sample consensus
64 recipes
65 recognition
66 recognition process
67 reconstruction
68 representation
69 respect
70 robot
71 robotic applications
72 robotics
73 sample consensus
74 second part
75 sensors
76 similarity
77 small spatial extent
78 solution
79 spatial extent
80 stereo problem
81 structure estimation
82 three-dimensional motion
83 time window
84 trajectories
85 two-dimensional images
86 vision
87 visual features
88 visual words
89 way
90 window
91 words
92 schema:name 3-D Vision and Recognition
93 schema:pagination 543-562
94 schema:productId N4e529a75f04945f984b16b33617fcdb4
95 N775ef1eab4ff431f8ae4ff1ffa18d332
96 schema:publisher N6b6e4cabefcc42738bb72863303cee21
97 schema:sameAs https://app.dimensions.ai/details/publication/pub.1029772224
98 https://doi.org/10.1007/978-3-540-30301-5_24
99 schema:sdDatePublished 2022-12-01T06:55
100 schema:sdLicense https://scigraph.springernature.com/explorer/license/
101 schema:sdPublisher Ncf0109e2264243f99d930f05ade7c5cd
102 schema:url https://doi.org/10.1007/978-3-540-30301-5_24
103 sgo:license sg:explorer/license/
104 sgo:sdDataset chapters
105 rdf:type schema:Chapter
106 N1b6cc67d2d174bc78e678b35cb9f41f7 schema:isbn 978-3-540-23957-4
107 978-3-540-30301-5
108 schema:name Springer Handbook of Robotics
109 rdf:type schema:Book
110 N328f078036634d6c8efcb394163bb4dc rdf:first sg:person.0643156134.22
111 rdf:rest N4867ce7ffdb243229007e259a8a94376
112 N4867ce7ffdb243229007e259a8a94376 rdf:first sg:person.014400652155.17
113 rdf:rest rdf:nil
114 N4e529a75f04945f984b16b33617fcdb4 schema:name doi
115 schema:value 10.1007/978-3-540-30301-5_24
116 rdf:type schema:PropertyValue
117 N6b6e4cabefcc42738bb72863303cee21 schema:name Springer Nature
118 rdf:type schema:Organisation
119 N6d66086b73794bb696a659e52e774208 schema:familyName Khatib
120 schema:givenName Oussama
121 rdf:type schema:Person
122 N775ef1eab4ff431f8ae4ff1ffa18d332 schema:name dimensions_id
123 schema:value pub.1029772224
124 rdf:type schema:PropertyValue
125 N858022c858244f6f81ef24ff40ac6daa rdf:first Nb88214b7822e4217b5513fc676feefa1
126 rdf:rest N8b219b7ebb9a4bc88e335ae50f221178
127 N8b219b7ebb9a4bc88e335ae50f221178 rdf:first N6d66086b73794bb696a659e52e774208
128 rdf:rest rdf:nil
129 Nb88214b7822e4217b5513fc676feefa1 schema:familyName Siciliano
130 schema:givenName Bruno
131 rdf:type schema:Person
132 Ncf0109e2264243f99d930f05ade7c5cd schema:name Springer Nature - SN SciGraph project
133 rdf:type schema:Organization
134 anzsrc-for:08 schema:inDefinedTermSet anzsrc-for:
135 schema:name Information and Computing Sciences
136 rdf:type schema:DefinedTerm
137 anzsrc-for:0801 schema:inDefinedTermSet anzsrc-for:
138 schema:name Artificial Intelligence and Image Processing
139 rdf:type schema:DefinedTerm
140 sg:person.014400652155.17 schema:affiliation grid-institutes:grid.5037.1
141 schema:familyName Eklundh
142 schema:givenName Jan-Olof
143 schema:sameAs https://app.dimensions.ai/discover/publication?and_facet_researcher=ur.014400652155.17
144 rdf:type schema:Person
145 sg:person.0643156134.22 schema:affiliation grid-institutes:grid.25879.31
146 schema:familyName Daniilidis
147 schema:givenName Kostas
148 schema:sameAs https://app.dimensions.ai/discover/publication?and_facet_researcher=ur.0643156134.22
149 rdf:type schema:Person
150 grid-institutes:grid.25879.31 schema:alternateName Department of Computer and Information Science GRASP Laboratory, University of Pennsylvania, 3330 Walnut Street, 19104, Philadelphia, PA, USA
151 schema:name Department of Computer and Information Science GRASP Laboratory, University of Pennsylvania, 3330 Walnut Street, 19104, Philadelphia, PA, USA
152 rdf:type schema:Organization
153 grid-institutes:grid.5037.1 schema:alternateName KTH Royal Institute of Technology, Teknikringen 14, 10044, Stockholm, Sweden
154 schema:name KTH Royal Institute of Technology, Teknikringen 14, 10044, Stockholm, Sweden
155 rdf:type schema:Organization
 




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


...