On the Significance of Real-World Conditions for Material Classification View Full Text


Ontology type: schema:Chapter      Open Access: True


Chapter Info

DATE

2004

AUTHORS

Eric Hayman , Barbara Caputo , Mario Fritz , Jan-Olof Eklundh

ABSTRACT

Classifying materials from their appearance is a challenging problem, especially if illumination and pose conditions are permitted to change: highlights and shadows caused by 3D structure can radically alter a sample’s visual texture. Despite these difficulties, researchers have demonstrated impressive results on the CUReT database which contains many images of 61 materials under different conditions. A first contribution of this paper is to further advance the state-of-the-art by applying Support Vector Machines to this problem. To our knowledge, we record the best results to date on the CUReT database.In our work we additionally investigate the effect of scale since robustness to viewing distance and zoom settings is crucial in many real-world situations. Indeed, a material’s appearance can vary considerably as fine-level detail becomes visible or disappears as the camera moves towards or away from the subject. We handle scale-variations using a pure-learning approach, incorporating samples imaged at different distances into the training set. An empirical investigation is conducted to show how the classification accuracy decreases as less scale information is made available during training. Since the CUReT database contains little scale variation, we introduce a new database which images ten CUReT materials at different distances, while also maintaining some change in pose and illumination. The first aim of the database is thus to provide scale variations, but a second and equally important objective is to attempt to recognise different samples of the CUReT materials. For instance, does training on the CUReT database enable recognition of another piece of sandpaper? The results clearly demonstrate that it is not possible to do so with any acceptable degree of accuracy. Thus we conclude that impressive results even on a well-designed database such as CUReT, does not imply that material classification is close to being a solved problem under real-world conditions. More... »

PAGES

253-266

Identifiers

URI

http://scigraph.springernature.com/pub.10.1007/978-3-540-24673-2_21

DOI

http://dx.doi.org/10.1007/978-3-540-24673-2_21

DIMENSIONS

https://app.dimensions.ai/details/publication/pub.1045827469


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": "Computational Vision and Active Perception Laboratory, Dept. of Numerical Analysis and Computer Science, Royal Institute of Technology (KTH), SE-100 44, Stockholm, Sweden", 
          "id": "http://www.grid.ac/institutes/grid.5037.1", 
          "name": [
            "Computational Vision and Active Perception Laboratory, Dept. of Numerical Analysis and Computer Science, Royal Institute of Technology (KTH), SE-100 44, Stockholm, Sweden"
          ], 
          "type": "Organization"
        }, 
        "familyName": "Hayman", 
        "givenName": "Eric", 
        "id": "sg:person.010203264647.00", 
        "sameAs": [
          "https://app.dimensions.ai/discover/publication?and_facet_researcher=ur.010203264647.00"
        ], 
        "type": "Person"
      }, 
      {
        "affiliation": {
          "alternateName": "Computational Vision and Active Perception Laboratory, Dept. of Numerical Analysis and Computer Science, Royal Institute of Technology (KTH), SE-100 44, Stockholm, Sweden", 
          "id": "http://www.grid.ac/institutes/grid.5037.1", 
          "name": [
            "Computational Vision and Active Perception Laboratory, Dept. of Numerical Analysis and Computer Science, Royal Institute of Technology (KTH), SE-100 44, Stockholm, Sweden"
          ], 
          "type": "Organization"
        }, 
        "familyName": "Caputo", 
        "givenName": "Barbara", 
        "id": "sg:person.012403660253.21", 
        "sameAs": [
          "https://app.dimensions.ai/discover/publication?and_facet_researcher=ur.012403660253.21"
        ], 
        "type": "Person"
      }, 
      {
        "affiliation": {
          "alternateName": "Computational Vision and Active Perception Laboratory, Dept. of Numerical Analysis and Computer Science, Royal Institute of Technology (KTH), SE-100 44, Stockholm, Sweden", 
          "id": "http://www.grid.ac/institutes/grid.5037.1", 
          "name": [
            "Computational Vision and Active Perception Laboratory, Dept. of Numerical Analysis and Computer Science, Royal Institute of Technology (KTH), SE-100 44, Stockholm, Sweden"
          ], 
          "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": "Computational Vision and Active Perception Laboratory, Dept. of Numerical Analysis and Computer Science, Royal Institute of Technology (KTH), SE-100 44, Stockholm, Sweden", 
          "id": "http://www.grid.ac/institutes/grid.5037.1", 
          "name": [
            "Computational Vision and Active Perception Laboratory, Dept. of Numerical Analysis and Computer Science, Royal Institute of Technology (KTH), SE-100 44, 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": "2004", 
    "datePublishedReg": "2004-01-01", 
    "description": "Classifying materials from their appearance is a challenging problem, especially if illumination and pose conditions are permitted to change: highlights and shadows caused by 3D structure can radically alter a sample\u2019s visual texture. Despite these difficulties, researchers have demonstrated impressive results on the CUReT database which contains many images of 61 materials under different conditions. A first contribution of this paper is to further advance the state-of-the-art by applying Support Vector Machines to this problem. To our knowledge, we record the best results to date on the CUReT database.In our work we additionally investigate the effect of scale since robustness to viewing distance and zoom settings is crucial in many real-world situations. Indeed, a material\u2019s appearance can vary considerably as fine-level detail becomes visible or disappears as the camera moves towards or away from the subject. We handle scale-variations using a pure-learning approach, incorporating samples imaged at different distances into the training set. An empirical investigation is conducted to show how the classification accuracy decreases as less scale information is made available during training. Since the CUReT database contains little scale variation, we introduce a new database which images ten CUReT materials at different distances, while also maintaining some change in pose and illumination. The first aim of the database is thus to provide scale variations, but a second and equally important objective is to attempt to recognise different samples of the CUReT materials. For instance, does training on the CUReT database enable recognition of another piece of sandpaper? The results clearly demonstrate that it is not possible to do so with any acceptable degree of accuracy. Thus we conclude that impressive results even on a well-designed database such as CUReT, does not imply that material classification is close to being a solved problem under real-world conditions.", 
    "editor": [
      {
        "familyName": "Pajdla", 
        "givenName": "Tom\u00e1s", 
        "type": "Person"
      }, 
      {
        "familyName": "Matas", 
        "givenName": "Ji\u0159\u00ed", 
        "type": "Person"
      }
    ], 
    "genre": "chapter", 
    "id": "sg:pub.10.1007/978-3-540-24673-2_21", 
    "isAccessibleForFree": true, 
    "isPartOf": {
      "isbn": [
        "978-3-540-21981-1", 
        "978-3-540-24673-2"
      ], 
      "name": "Computer Vision - ECCV 2004", 
      "type": "Book"
    }, 
    "keywords": [
      "CUReT database", 
      "pure learning approach", 
      "classification accuracy decreases", 
      "impressive results", 
      "support vector machine", 
      "material classification", 
      "fine-level details", 
      "scale variations", 
      "visual texture", 
      "real-world situations", 
      "pose conditions", 
      "vector machine", 
      "real-world conditions", 
      "challenging problem", 
      "training set", 
      "scale information", 
      "material appearance", 
      "piece of sandpaper", 
      "first contribution", 
      "world conditions", 
      "accuracy decreases", 
      "new database", 
      "database", 
      "classification", 
      "better results", 
      "enable recognition", 
      "pose", 
      "important objective", 
      "machine", 
      "camera", 
      "images", 
      "robustness", 
      "training", 
      "recognition", 
      "accuracy", 
      "acceptable degree", 
      "information", 
      "set", 
      "instances", 
      "art", 
      "different distances", 
      "empirical investigation", 
      "curet", 
      "illumination", 
      "researchers", 
      "distance", 
      "shadow", 
      "results", 
      "texture", 
      "knowledge", 
      "work", 
      "situation", 
      "pieces", 
      "detail", 
      "difficulties", 
      "different conditions", 
      "effect of scale", 
      "objective", 
      "state", 
      "highlights", 
      "setting", 
      "appearance", 
      "contribution", 
      "structure", 
      "first aim", 
      "scale", 
      "conditions", 
      "degree", 
      "date", 
      "variation", 
      "aim", 
      "different samples", 
      "significance", 
      "subjects", 
      "changes", 
      "investigation", 
      "samples", 
      "materials", 
      "sandpaper", 
      "effect", 
      "decrease", 
      "problem", 
      "paper", 
      "approach"
    ], 
    "name": "On the Significance of Real-World Conditions for Material Classification", 
    "pagination": "253-266", 
    "productId": [
      {
        "name": "dimensions_id", 
        "type": "PropertyValue", 
        "value": [
          "pub.1045827469"
        ]
      }, 
      {
        "name": "doi", 
        "type": "PropertyValue", 
        "value": [
          "10.1007/978-3-540-24673-2_21"
        ]
      }
    ], 
    "publisher": {
      "name": "Springer Nature", 
      "type": "Organisation"
    }, 
    "sameAs": [
      "https://doi.org/10.1007/978-3-540-24673-2_21", 
      "https://app.dimensions.ai/details/publication/pub.1045827469"
    ], 
    "sdDataset": "chapters", 
    "sdDatePublished": "2022-08-04T17:18", 
    "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/chapter/chapter_284.jsonl", 
    "type": "Chapter", 
    "url": "https://doi.org/10.1007/978-3-540-24673-2_21"
  }
]
 

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.10.1007/978-3-540-24673-2_21'

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

curl -H 'Accept: application/n-triples' 'https://scigraph.springernature.com/pub.10.1007/978-3-540-24673-2_21'

Turtle is a human-readable linked data format.

curl -H 'Accept: text/turtle' 'https://scigraph.springernature.com/pub.10.1007/978-3-540-24673-2_21'

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

curl -H 'Accept: application/rdf+xml' 'https://scigraph.springernature.com/pub.10.1007/978-3-540-24673-2_21'


 

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

169 TRIPLES      22 PREDICATES      109 URIs      102 LITERALS      7 BLANK NODES

Subject Predicate Object
1 sg:pub.10.1007/978-3-540-24673-2_21 schema:about anzsrc-for:08
2 anzsrc-for:0801
3 schema:author N829687dd415f4db1a2af3543ddcdeae3
4 schema:datePublished 2004
5 schema:datePublishedReg 2004-01-01
6 schema:description Classifying materials from their appearance is a challenging problem, especially if illumination and pose conditions are permitted to change: highlights and shadows caused by 3D structure can radically alter a sample’s visual texture. Despite these difficulties, researchers have demonstrated impressive results on the CUReT database which contains many images of 61 materials under different conditions. A first contribution of this paper is to further advance the state-of-the-art by applying Support Vector Machines to this problem. To our knowledge, we record the best results to date on the CUReT database.In our work we additionally investigate the effect of scale since robustness to viewing distance and zoom settings is crucial in many real-world situations. Indeed, a material’s appearance can vary considerably as fine-level detail becomes visible or disappears as the camera moves towards or away from the subject. We handle scale-variations using a pure-learning approach, incorporating samples imaged at different distances into the training set. An empirical investigation is conducted to show how the classification accuracy decreases as less scale information is made available during training. Since the CUReT database contains little scale variation, we introduce a new database which images ten CUReT materials at different distances, while also maintaining some change in pose and illumination. The first aim of the database is thus to provide scale variations, but a second and equally important objective is to attempt to recognise different samples of the CUReT materials. For instance, does training on the CUReT database enable recognition of another piece of sandpaper? The results clearly demonstrate that it is not possible to do so with any acceptable degree of accuracy. Thus we conclude that impressive results even on a well-designed database such as CUReT, does not imply that material classification is close to being a solved problem under real-world conditions.
7 schema:editor Nf81c3d6867ff4fd984a9d9e35e5af408
8 schema:genre chapter
9 schema:isAccessibleForFree true
10 schema:isPartOf Nb133ec8295bc4f71b4a22e7eb33088e8
11 schema:keywords CUReT database
12 acceptable degree
13 accuracy
14 accuracy decreases
15 aim
16 appearance
17 approach
18 art
19 better results
20 camera
21 challenging problem
22 changes
23 classification
24 classification accuracy decreases
25 conditions
26 contribution
27 curet
28 database
29 date
30 decrease
31 degree
32 detail
33 different conditions
34 different distances
35 different samples
36 difficulties
37 distance
38 effect
39 effect of scale
40 empirical investigation
41 enable recognition
42 fine-level details
43 first aim
44 first contribution
45 highlights
46 illumination
47 images
48 important objective
49 impressive results
50 information
51 instances
52 investigation
53 knowledge
54 machine
55 material appearance
56 material classification
57 materials
58 new database
59 objective
60 paper
61 piece of sandpaper
62 pieces
63 pose
64 pose conditions
65 problem
66 pure learning approach
67 real-world conditions
68 real-world situations
69 recognition
70 researchers
71 results
72 robustness
73 samples
74 sandpaper
75 scale
76 scale information
77 scale variations
78 set
79 setting
80 shadow
81 significance
82 situation
83 state
84 structure
85 subjects
86 support vector machine
87 texture
88 training
89 training set
90 variation
91 vector machine
92 visual texture
93 work
94 world conditions
95 schema:name On the Significance of Real-World Conditions for Material Classification
96 schema:pagination 253-266
97 schema:productId Nacd6f17545374d15a506cfe1d0b6b040
98 Nea38f73d42a74056a0f3a525c6c5d460
99 schema:publisher N6dd2d698158747a485cc44d084b69832
100 schema:sameAs https://app.dimensions.ai/details/publication/pub.1045827469
101 https://doi.org/10.1007/978-3-540-24673-2_21
102 schema:sdDatePublished 2022-08-04T17:18
103 schema:sdLicense https://scigraph.springernature.com/explorer/license/
104 schema:sdPublisher N47721632c292436a91d64dadf4100c2c
105 schema:url https://doi.org/10.1007/978-3-540-24673-2_21
106 sgo:license sg:explorer/license/
107 sgo:sdDataset chapters
108 rdf:type schema:Chapter
109 N47721632c292436a91d64dadf4100c2c schema:name Springer Nature - SN SciGraph project
110 rdf:type schema:Organization
111 N5b52d77c07234b5ea5c5907fed5a3126 rdf:first sg:person.012403660253.21
112 rdf:rest N8a24ebc94722496e81e5f17476032731
113 N6dd2d698158747a485cc44d084b69832 schema:name Springer Nature
114 rdf:type schema:Organisation
115 N829687dd415f4db1a2af3543ddcdeae3 rdf:first sg:person.010203264647.00
116 rdf:rest N5b52d77c07234b5ea5c5907fed5a3126
117 N8a24ebc94722496e81e5f17476032731 rdf:first sg:person.013361072755.17
118 rdf:rest Nfdecaec1c040465f9d756c5e734f81ca
119 N9d0842fb3e444977a67ea79057acd915 rdf:first Nfef0658d2e494f10937f0d18c7a1dfaa
120 rdf:rest rdf:nil
121 Nacd6f17545374d15a506cfe1d0b6b040 schema:name doi
122 schema:value 10.1007/978-3-540-24673-2_21
123 rdf:type schema:PropertyValue
124 Nb133ec8295bc4f71b4a22e7eb33088e8 schema:isbn 978-3-540-21981-1
125 978-3-540-24673-2
126 schema:name Computer Vision - ECCV 2004
127 rdf:type schema:Book
128 Nc27101f9d82a4b90af35449793f15139 schema:familyName Pajdla
129 schema:givenName Tomás
130 rdf:type schema:Person
131 Nea38f73d42a74056a0f3a525c6c5d460 schema:name dimensions_id
132 schema:value pub.1045827469
133 rdf:type schema:PropertyValue
134 Nf81c3d6867ff4fd984a9d9e35e5af408 rdf:first Nc27101f9d82a4b90af35449793f15139
135 rdf:rest N9d0842fb3e444977a67ea79057acd915
136 Nfdecaec1c040465f9d756c5e734f81ca rdf:first sg:person.014400652155.17
137 rdf:rest rdf:nil
138 Nfef0658d2e494f10937f0d18c7a1dfaa schema:familyName Matas
139 schema:givenName Jiří
140 rdf:type schema:Person
141 anzsrc-for:08 schema:inDefinedTermSet anzsrc-for:
142 schema:name Information and Computing Sciences
143 rdf:type schema:DefinedTerm
144 anzsrc-for:0801 schema:inDefinedTermSet anzsrc-for:
145 schema:name Artificial Intelligence and Image Processing
146 rdf:type schema:DefinedTerm
147 sg:person.010203264647.00 schema:affiliation grid-institutes:grid.5037.1
148 schema:familyName Hayman
149 schema:givenName Eric
150 schema:sameAs https://app.dimensions.ai/discover/publication?and_facet_researcher=ur.010203264647.00
151 rdf:type schema:Person
152 sg:person.012403660253.21 schema:affiliation grid-institutes:grid.5037.1
153 schema:familyName Caputo
154 schema:givenName Barbara
155 schema:sameAs https://app.dimensions.ai/discover/publication?and_facet_researcher=ur.012403660253.21
156 rdf:type schema:Person
157 sg:person.013361072755.17 schema:affiliation grid-institutes:grid.5037.1
158 schema:familyName Fritz
159 schema:givenName Mario
160 schema:sameAs https://app.dimensions.ai/discover/publication?and_facet_researcher=ur.013361072755.17
161 rdf:type schema:Person
162 sg:person.014400652155.17 schema:affiliation grid-institutes:grid.5037.1
163 schema:familyName Eklundh
164 schema:givenName Jan-Olof
165 schema:sameAs https://app.dimensions.ai/discover/publication?and_facet_researcher=ur.014400652155.17
166 rdf:type schema:Person
167 grid-institutes:grid.5037.1 schema:alternateName Computational Vision and Active Perception Laboratory, Dept. of Numerical Analysis and Computer Science, Royal Institute of Technology (KTH), SE-100 44, Stockholm, Sweden
168 schema:name Computational Vision and Active Perception Laboratory, Dept. of Numerical Analysis and Computer Science, Royal Institute of Technology (KTH), SE-100 44, Stockholm, Sweden
169 rdf:type schema:Organization
 




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


...