Synthetic Convolutional Features for Improved Semantic Segmentation View Full Text


Ontology type: schema:Chapter      Open Access: True


Chapter Info

DATE

2020

AUTHORS

Yang He , Bernt Schiele , Mario Fritz

ABSTRACT

Recently, learning-based image synthesis has enabled to generate high resolution images, either applying popular adversarial training or a powerful perceptual loss. However, it remains challenging to successfully leverage synthetic data for improving semantic segmentation with additional synthetic images. Therefore, we suggest to generate intermediate convolutional features and propose the first synthesis approach that is catered to such intermediate convolutional features. This allows us to generate new features from label masks and include them successfully into the training procedure in order to improve the performance of semantic segmentation. Experimental results and analysis on two challenging datasets Cityscapes and ADE20K show that our generated feature improves performance on segmentation tasks. More... »

PAGES

320-336

Identifiers

URI

http://scigraph.springernature.com/pub.10.1007/978-3-030-66823-5_19

DOI

http://dx.doi.org/10.1007/978-3-030-66823-5_19

DIMENSIONS

https://app.dimensions.ai/details/publication/pub.1134282973


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": "Max Planck Institute for Informatics, Saarland Informatics Campus, Saarbr\u00fccken, Germany", 
          "id": "http://www.grid.ac/institutes/grid.419528.3", 
          "name": [
            "CISPA Helmholtz Center for Information Security, Saarbr\u00fccken, Germany", 
            "Max Planck Institute for Informatics, Saarland Informatics Campus, Saarbr\u00fccken, Germany"
          ], 
          "type": "Organization"
        }, 
        "familyName": "He", 
        "givenName": "Yang", 
        "id": "sg:person.010655401332.41", 
        "sameAs": [
          "https://app.dimensions.ai/discover/publication?and_facet_researcher=ur.010655401332.41"
        ], 
        "type": "Person"
      }, 
      {
        "affiliation": {
          "alternateName": "Max Planck Institute for Informatics, Saarland Informatics Campus, Saarbr\u00fccken, Germany", 
          "id": "http://www.grid.ac/institutes/grid.419528.3", 
          "name": [
            "Max Planck Institute for Informatics, Saarland Informatics Campus, Saarbr\u00fccken, Germany"
          ], 
          "type": "Organization"
        }, 
        "familyName": "Schiele", 
        "givenName": "Bernt", 
        "id": "sg:person.01174260421.90", 
        "sameAs": [
          "https://app.dimensions.ai/discover/publication?and_facet_researcher=ur.01174260421.90"
        ], 
        "type": "Person"
      }, 
      {
        "affiliation": {
          "alternateName": "CISPA Helmholtz Center for Information Security, Saarbr\u00fccken, Germany", 
          "id": "http://www.grid.ac/institutes/grid.507511.7", 
          "name": [
            "CISPA Helmholtz Center for Information Security, 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"
      }
    ], 
    "datePublished": "2020", 
    "datePublishedReg": "2020-01-01", 
    "description": "Recently, learning-based image synthesis has enabled to generate high resolution images, either applying popular adversarial training or a powerful perceptual loss. However, it remains challenging to successfully leverage synthetic data for improving semantic segmentation with additional synthetic images. Therefore, we suggest to generate intermediate convolutional features and propose the first synthesis approach that is catered to such intermediate convolutional features. This allows us to generate new features from label masks and include them successfully into the training procedure in order to improve the performance of semantic segmentation. Experimental results and analysis on two challenging datasets Cityscapes and ADE20K show that our generated feature improves performance on segmentation tasks.", 
    "editor": [
      {
        "familyName": "Bartoli", 
        "givenName": "Adrien", 
        "type": "Person"
      }, 
      {
        "familyName": "Fusiello", 
        "givenName": "Andrea", 
        "type": "Person"
      }
    ], 
    "genre": "chapter", 
    "id": "sg:pub.10.1007/978-3-030-66823-5_19", 
    "isAccessibleForFree": true, 
    "isPartOf": {
      "isbn": [
        "978-3-030-66822-8", 
        "978-3-030-66823-5"
      ], 
      "name": "Computer Vision \u2013 ECCV 2020 Workshops", 
      "type": "Book"
    }, 
    "keywords": [
      "semantic segmentation", 
      "convolutional features", 
      "segmentation task", 
      "image synthesis", 
      "adversarial training", 
      "perceptual loss", 
      "ADE20K", 
      "high-resolution images", 
      "synthetic images", 
      "label mask", 
      "synthetic data", 
      "segmentation", 
      "resolution images", 
      "new features", 
      "training procedure", 
      "experimental results", 
      "images", 
      "features", 
      "synthesis approach", 
      "performance", 
      "cityscape", 
      "task", 
      "training", 
      "mask", 
      "data", 
      "order", 
      "results", 
      "procedure", 
      "analysis", 
      "loss", 
      "synthesis", 
      "approach"
    ], 
    "name": "Synthetic Convolutional Features for Improved Semantic Segmentation", 
    "pagination": "320-336", 
    "productId": [
      {
        "name": "dimensions_id", 
        "type": "PropertyValue", 
        "value": [
          "pub.1134282973"
        ]
      }, 
      {
        "name": "doi", 
        "type": "PropertyValue", 
        "value": [
          "10.1007/978-3-030-66823-5_19"
        ]
      }
    ], 
    "publisher": {
      "name": "Springer Nature", 
      "type": "Organisation"
    }, 
    "sameAs": [
      "https://doi.org/10.1007/978-3-030-66823-5_19", 
      "https://app.dimensions.ai/details/publication/pub.1134282973"
    ], 
    "sdDataset": "chapters", 
    "sdDatePublished": "2022-11-24T21:13", 
    "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_206.jsonl", 
    "type": "Chapter", 
    "url": "https://doi.org/10.1007/978-3-030-66823-5_19"
  }
]
 

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.10.1007/978-3-030-66823-5_19'

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

curl -H 'Accept: application/n-triples' 'https://scigraph.springernature.com/pub.10.1007/978-3-030-66823-5_19'

Turtle is a human-readable linked data format.

curl -H 'Accept: text/turtle' 'https://scigraph.springernature.com/pub.10.1007/978-3-030-66823-5_19'

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

curl -H 'Accept: application/rdf+xml' 'https://scigraph.springernature.com/pub.10.1007/978-3-030-66823-5_19'


 

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

114 TRIPLES      22 PREDICATES      57 URIs      50 LITERALS      7 BLANK NODES

Subject Predicate Object
1 sg:pub.10.1007/978-3-030-66823-5_19 schema:about anzsrc-for:08
2 anzsrc-for:0801
3 schema:author Na62e425dd83c460ebb1e6739e033e643
4 schema:datePublished 2020
5 schema:datePublishedReg 2020-01-01
6 schema:description Recently, learning-based image synthesis has enabled to generate high resolution images, either applying popular adversarial training or a powerful perceptual loss. However, it remains challenging to successfully leverage synthetic data for improving semantic segmentation with additional synthetic images. Therefore, we suggest to generate intermediate convolutional features and propose the first synthesis approach that is catered to such intermediate convolutional features. This allows us to generate new features from label masks and include them successfully into the training procedure in order to improve the performance of semantic segmentation. Experimental results and analysis on two challenging datasets Cityscapes and ADE20K show that our generated feature improves performance on segmentation tasks.
7 schema:editor Ne6d67658b301487680e70904a8102dd2
8 schema:genre chapter
9 schema:isAccessibleForFree true
10 schema:isPartOf N92ef4a81d4494bba8289cbf6f87f2cb7
11 schema:keywords ADE20K
12 adversarial training
13 analysis
14 approach
15 cityscape
16 convolutional features
17 data
18 experimental results
19 features
20 high-resolution images
21 image synthesis
22 images
23 label mask
24 loss
25 mask
26 new features
27 order
28 perceptual loss
29 performance
30 procedure
31 resolution images
32 results
33 segmentation
34 segmentation task
35 semantic segmentation
36 synthesis
37 synthesis approach
38 synthetic data
39 synthetic images
40 task
41 training
42 training procedure
43 schema:name Synthetic Convolutional Features for Improved Semantic Segmentation
44 schema:pagination 320-336
45 schema:productId N12f5e4a30a2747c7bb210a1ba9ea97ed
46 Nf0758df695ee413cbfa4a7b2330296e7
47 schema:publisher Nfed3cf3763614121b256db9a4908ecea
48 schema:sameAs https://app.dimensions.ai/details/publication/pub.1134282973
49 https://doi.org/10.1007/978-3-030-66823-5_19
50 schema:sdDatePublished 2022-11-24T21:13
51 schema:sdLicense https://scigraph.springernature.com/explorer/license/
52 schema:sdPublisher N0b2bf3079a334d4abd627bc2c8bed938
53 schema:url https://doi.org/10.1007/978-3-030-66823-5_19
54 sgo:license sg:explorer/license/
55 sgo:sdDataset chapters
56 rdf:type schema:Chapter
57 N0b2bf3079a334d4abd627bc2c8bed938 schema:name Springer Nature - SN SciGraph project
58 rdf:type schema:Organization
59 N12f5e4a30a2747c7bb210a1ba9ea97ed schema:name dimensions_id
60 schema:value pub.1134282973
61 rdf:type schema:PropertyValue
62 N17285166bf114ec1956c93acd0ce86ca rdf:first sg:person.013361072755.17
63 rdf:rest rdf:nil
64 N3c928a645c2645aea575742e022da7fb rdf:first sg:person.01174260421.90
65 rdf:rest N17285166bf114ec1956c93acd0ce86ca
66 N788bd0a40bdf4ab8a8218243546e3118 schema:familyName Bartoli
67 schema:givenName Adrien
68 rdf:type schema:Person
69 N80c93281863b4c72b681a9a325801d0a rdf:first Nbb216d2af71a475ab46364fb591a278f
70 rdf:rest rdf:nil
71 N92ef4a81d4494bba8289cbf6f87f2cb7 schema:isbn 978-3-030-66822-8
72 978-3-030-66823-5
73 schema:name Computer Vision – ECCV 2020 Workshops
74 rdf:type schema:Book
75 Na62e425dd83c460ebb1e6739e033e643 rdf:first sg:person.010655401332.41
76 rdf:rest N3c928a645c2645aea575742e022da7fb
77 Nbb216d2af71a475ab46364fb591a278f schema:familyName Fusiello
78 schema:givenName Andrea
79 rdf:type schema:Person
80 Ne6d67658b301487680e70904a8102dd2 rdf:first N788bd0a40bdf4ab8a8218243546e3118
81 rdf:rest N80c93281863b4c72b681a9a325801d0a
82 Nf0758df695ee413cbfa4a7b2330296e7 schema:name doi
83 schema:value 10.1007/978-3-030-66823-5_19
84 rdf:type schema:PropertyValue
85 Nfed3cf3763614121b256db9a4908ecea schema:name Springer Nature
86 rdf:type schema:Organisation
87 anzsrc-for:08 schema:inDefinedTermSet anzsrc-for:
88 schema:name Information and Computing Sciences
89 rdf:type schema:DefinedTerm
90 anzsrc-for:0801 schema:inDefinedTermSet anzsrc-for:
91 schema:name Artificial Intelligence and Image Processing
92 rdf:type schema:DefinedTerm
93 sg:person.010655401332.41 schema:affiliation grid-institutes:grid.419528.3
94 schema:familyName He
95 schema:givenName Yang
96 schema:sameAs https://app.dimensions.ai/discover/publication?and_facet_researcher=ur.010655401332.41
97 rdf:type schema:Person
98 sg:person.01174260421.90 schema:affiliation grid-institutes:grid.419528.3
99 schema:familyName Schiele
100 schema:givenName Bernt
101 schema:sameAs https://app.dimensions.ai/discover/publication?and_facet_researcher=ur.01174260421.90
102 rdf:type schema:Person
103 sg:person.013361072755.17 schema:affiliation grid-institutes:grid.507511.7
104 schema:familyName Fritz
105 schema:givenName Mario
106 schema:sameAs https://app.dimensions.ai/discover/publication?and_facet_researcher=ur.013361072755.17
107 rdf:type schema:Person
108 grid-institutes:grid.419528.3 schema:alternateName Max Planck Institute for Informatics, Saarland Informatics Campus, Saarbrücken, Germany
109 schema:name CISPA Helmholtz Center for Information Security, Saarbrücken, Germany
110 Max Planck Institute for Informatics, Saarland Informatics Campus, Saarbrücken, Germany
111 rdf:type schema:Organization
112 grid-institutes:grid.507511.7 schema:alternateName CISPA Helmholtz Center for Information Security, Saarbrücken, Germany
113 schema:name CISPA Helmholtz Center for Information Security, Saarbrücken, Germany
114 rdf:type schema:Organization
 




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


...