Body Shape Privacy in Images: Understanding Privacy and Preventing Automatic Shape Extraction View Full Text


Ontology type: schema:Chapter      Open Access: True


Chapter Info

DATE

2020

AUTHORS

Hosnieh Sattar , Katharina Krombholz , Gerard Pons-Moll , Mario Fritz

ABSTRACT

Modern approaches to pose and body shape estimation have recently achieved strong performance even under challenging real-world conditions. Even from a single image of a clothed person, a realistic looking body shape can be inferred that captures a users’ weight group and body shape type well. This opens up a whole spectrum of applications – in particular in fashion – where virtual try-on and recommendation systems can make use of these new and automatized cues. However, a realistic depiction of the undressed body is regarded highly private and therefore might not be consented by most people. Hence, we ask if the automatic extraction of such information can be effectively evaded. While adversarial perturbations have been shown to be effective for manipulating the output of machine learning models – in particular, end-to-end deep learning approaches – state of the art shape estimation methods are composed of multiple stages. We perform the first investigation of different strategies that can be used to effectively manipulate the automatic shape estimation while preserving the overall appearance of the original image. More... »

PAGES

411-428

Identifiers

URI

http://scigraph.springernature.com/pub.10.1007/978-3-030-68238-5_31

DOI

http://dx.doi.org/10.1007/978-3-030-68238-5_31

DIMENSIONS

https://app.dimensions.ai/details/publication/pub.1135004839


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, Saarbr\u00fccken, Germany", 
          "id": "http://www.grid.ac/institutes/grid.419528.3", 
          "name": [
            "Max Planck Institute for Informatics, Saarbr\u00fccken, Germany"
          ], 
          "type": "Organization"
        }, 
        "familyName": "Sattar", 
        "givenName": "Hosnieh", 
        "id": "sg:person.013434275371.03", 
        "sameAs": [
          "https://app.dimensions.ai/discover/publication?and_facet_researcher=ur.013434275371.03"
        ], 
        "type": "Person"
      }, 
      {
        "affiliation": {
          "alternateName": "CISPA Helmholtz Center for Information Security, Saarland Informatics Campus, Saarbr\u00fccken, Germany", 
          "id": "http://www.grid.ac/institutes/grid.507511.7", 
          "name": [
            "CISPA Helmholtz Center for Information Security, Saarland Informatics Campus, Saarbr\u00fccken, Germany"
          ], 
          "type": "Organization"
        }, 
        "familyName": "Krombholz", 
        "givenName": "Katharina", 
        "id": "sg:person.015614373645.33", 
        "sameAs": [
          "https://app.dimensions.ai/discover/publication?and_facet_researcher=ur.015614373645.33"
        ], 
        "type": "Person"
      }, 
      {
        "affiliation": {
          "alternateName": "Max Planck Institute for Informatics, Saarbr\u00fccken, Germany", 
          "id": "http://www.grid.ac/institutes/grid.419528.3", 
          "name": [
            "Max Planck Institute for Informatics, Saarbr\u00fccken, Germany"
          ], 
          "type": "Organization"
        }, 
        "familyName": "Pons-Moll", 
        "givenName": "Gerard", 
        "id": "sg:person.015242465107.59", 
        "sameAs": [
          "https://app.dimensions.ai/discover/publication?and_facet_researcher=ur.015242465107.59"
        ], 
        "type": "Person"
      }, 
      {
        "affiliation": {
          "alternateName": "CISPA Helmholtz Center for Information Security, Saarland Informatics Campus, Saarbr\u00fccken, Germany", 
          "id": "http://www.grid.ac/institutes/grid.507511.7", 
          "name": [
            "CISPA Helmholtz Center for Information Security, Saarland Informatics Campus, 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": "Modern approaches to pose and body shape estimation have recently achieved strong performance even under challenging real-world conditions. Even from a single image of a clothed person, a realistic looking body shape can be inferred that captures a users\u2019 weight group and body shape type well. This opens up a whole spectrum of applications \u2013 in particular in fashion \u2013 where virtual try-on and recommendation systems can make use of these new and automatized cues. However, a realistic depiction of the undressed body is regarded highly private and therefore might not be consented by most people. Hence, we ask if the automatic extraction of such information can be effectively evaded. While adversarial perturbations have been shown to be effective for manipulating the output of machine learning models \u2013 in particular, end-to-end deep learning approaches \u2013 state of the art shape estimation methods are composed of multiple stages. We perform the first investigation of different strategies that can be used to effectively manipulate the automatic shape estimation while preserving the overall appearance of the original image.", 
    "editor": [
      {
        "familyName": "Bartoli", 
        "givenName": "Adrien", 
        "type": "Person"
      }, 
      {
        "familyName": "Fusiello", 
        "givenName": "Andrea", 
        "type": "Person"
      }
    ], 
    "genre": "chapter", 
    "id": "sg:pub.10.1007/978-3-030-68238-5_31", 
    "isAccessibleForFree": true, 
    "isPartOf": {
      "isbn": [
        "978-3-030-68237-8", 
        "978-3-030-68238-5"
      ], 
      "name": "Computer Vision \u2013 ECCV 2020 Workshops", 
      "type": "Book"
    }, 
    "keywords": [
      "shape estimation", 
      "end deep learning", 
      "body shape estimation", 
      "output of machine", 
      "deep learning", 
      "recommendation system", 
      "adversarial perturbations", 
      "automatic extraction", 
      "single image", 
      "original image", 
      "shape extraction", 
      "real-world conditions", 
      "shape estimation method", 
      "undressed body", 
      "Understanding Privacy", 
      "privacy", 
      "strong performance", 
      "images", 
      "such information", 
      "estimation method", 
      "users", 
      "machine", 
      "modern approaches", 
      "extraction", 
      "multiple stages", 
      "learning", 
      "shape types", 
      "different strategies", 
      "estimation", 
      "information", 
      "applications", 
      "performance", 
      "realistic depiction", 
      "most people", 
      "system", 
      "body shape", 
      "output", 
      "model", 
      "whole spectrum", 
      "method", 
      "fashion", 
      "strategies", 
      "overall appearance", 
      "end", 
      "use", 
      "people", 
      "cues", 
      "persons", 
      "shape", 
      "types", 
      "first investigation", 
      "stage", 
      "depiction", 
      "appearance", 
      "body", 
      "conditions", 
      "perturbations", 
      "investigation", 
      "spectra", 
      "group", 
      "weight group", 
      "approach"
    ], 
    "name": "Body Shape Privacy in Images: Understanding Privacy and Preventing Automatic Shape Extraction", 
    "pagination": "411-428", 
    "productId": [
      {
        "name": "dimensions_id", 
        "type": "PropertyValue", 
        "value": [
          "pub.1135004839"
        ]
      }, 
      {
        "name": "doi", 
        "type": "PropertyValue", 
        "value": [
          "10.1007/978-3-030-68238-5_31"
        ]
      }
    ], 
    "publisher": {
      "name": "Springer Nature", 
      "type": "Organisation"
    }, 
    "sameAs": [
      "https://doi.org/10.1007/978-3-030-68238-5_31", 
      "https://app.dimensions.ai/details/publication/pub.1135004839"
    ], 
    "sdDataset": "chapters", 
    "sdDatePublished": "2022-11-24T21:15", 
    "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_305.jsonl", 
    "type": "Chapter", 
    "url": "https://doi.org/10.1007/978-3-030-68238-5_31"
  }
]
 

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.10.1007/978-3-030-68238-5_31'

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

curl -H 'Accept: application/n-triples' 'https://scigraph.springernature.com/pub.10.1007/978-3-030-68238-5_31'

Turtle is a human-readable linked data format.

curl -H 'Accept: text/turtle' 'https://scigraph.springernature.com/pub.10.1007/978-3-030-68238-5_31'

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

curl -H 'Accept: application/rdf+xml' 'https://scigraph.springernature.com/pub.10.1007/978-3-030-68238-5_31'


 

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

150 TRIPLES      22 PREDICATES      87 URIs      80 LITERALS      7 BLANK NODES

Subject Predicate Object
1 sg:pub.10.1007/978-3-030-68238-5_31 schema:about anzsrc-for:08
2 anzsrc-for:0801
3 schema:author N89fbd2c8a9f0433597032f2a2caf077c
4 schema:datePublished 2020
5 schema:datePublishedReg 2020-01-01
6 schema:description Modern approaches to pose and body shape estimation have recently achieved strong performance even under challenging real-world conditions. Even from a single image of a clothed person, a realistic looking body shape can be inferred that captures a users’ weight group and body shape type well. This opens up a whole spectrum of applications – in particular in fashion – where virtual try-on and recommendation systems can make use of these new and automatized cues. However, a realistic depiction of the undressed body is regarded highly private and therefore might not be consented by most people. Hence, we ask if the automatic extraction of such information can be effectively evaded. While adversarial perturbations have been shown to be effective for manipulating the output of machine learning models – in particular, end-to-end deep learning approaches – state of the art shape estimation methods are composed of multiple stages. We perform the first investigation of different strategies that can be used to effectively manipulate the automatic shape estimation while preserving the overall appearance of the original image.
7 schema:editor N86127c949f054121b806c955e5fbe48d
8 schema:genre chapter
9 schema:isAccessibleForFree true
10 schema:isPartOf N1f114f8486f3455590bf8054c69ef1ea
11 schema:keywords Understanding Privacy
12 adversarial perturbations
13 appearance
14 applications
15 approach
16 automatic extraction
17 body
18 body shape
19 body shape estimation
20 conditions
21 cues
22 deep learning
23 depiction
24 different strategies
25 end
26 end deep learning
27 estimation
28 estimation method
29 extraction
30 fashion
31 first investigation
32 group
33 images
34 information
35 investigation
36 learning
37 machine
38 method
39 model
40 modern approaches
41 most people
42 multiple stages
43 original image
44 output
45 output of machine
46 overall appearance
47 people
48 performance
49 persons
50 perturbations
51 privacy
52 real-world conditions
53 realistic depiction
54 recommendation system
55 shape
56 shape estimation
57 shape estimation method
58 shape extraction
59 shape types
60 single image
61 spectra
62 stage
63 strategies
64 strong performance
65 such information
66 system
67 types
68 undressed body
69 use
70 users
71 weight group
72 whole spectrum
73 schema:name Body Shape Privacy in Images: Understanding Privacy and Preventing Automatic Shape Extraction
74 schema:pagination 411-428
75 schema:productId N62f872066fb6487e9965827686270dbf
76 Nf41d15e2289844819c205576b0af178c
77 schema:publisher N46011c64b06c4ab9a4dac9fb2eda2e8a
78 schema:sameAs https://app.dimensions.ai/details/publication/pub.1135004839
79 https://doi.org/10.1007/978-3-030-68238-5_31
80 schema:sdDatePublished 2022-11-24T21:15
81 schema:sdLicense https://scigraph.springernature.com/explorer/license/
82 schema:sdPublisher N129c2c9428db4aedbe1c17e4b72386ba
83 schema:url https://doi.org/10.1007/978-3-030-68238-5_31
84 sgo:license sg:explorer/license/
85 sgo:sdDataset chapters
86 rdf:type schema:Chapter
87 N060e6049c759480b9555d75e215ba0e7 rdf:first sg:person.015614373645.33
88 rdf:rest N9d79ee5354cb4f4d80177f9fea8ca98f
89 N081fa8f936fc48c688dad0b7ded263f3 schema:familyName Fusiello
90 schema:givenName Andrea
91 rdf:type schema:Person
92 N0f5646be20214aaeab651660dfba1165 schema:familyName Bartoli
93 schema:givenName Adrien
94 rdf:type schema:Person
95 N129c2c9428db4aedbe1c17e4b72386ba schema:name Springer Nature - SN SciGraph project
96 rdf:type schema:Organization
97 N1f114f8486f3455590bf8054c69ef1ea schema:isbn 978-3-030-68237-8
98 978-3-030-68238-5
99 schema:name Computer Vision – ECCV 2020 Workshops
100 rdf:type schema:Book
101 N46011c64b06c4ab9a4dac9fb2eda2e8a schema:name Springer Nature
102 rdf:type schema:Organisation
103 N62f872066fb6487e9965827686270dbf schema:name dimensions_id
104 schema:value pub.1135004839
105 rdf:type schema:PropertyValue
106 N86127c949f054121b806c955e5fbe48d rdf:first N0f5646be20214aaeab651660dfba1165
107 rdf:rest N89aee0fe67f54fe391644c5d384a56b3
108 N89aee0fe67f54fe391644c5d384a56b3 rdf:first N081fa8f936fc48c688dad0b7ded263f3
109 rdf:rest rdf:nil
110 N89fbd2c8a9f0433597032f2a2caf077c rdf:first sg:person.013434275371.03
111 rdf:rest N060e6049c759480b9555d75e215ba0e7
112 N9d79ee5354cb4f4d80177f9fea8ca98f rdf:first sg:person.015242465107.59
113 rdf:rest Nf13f2a8b7e854a8fb7b73f014e3ef1ad
114 Nf13f2a8b7e854a8fb7b73f014e3ef1ad rdf:first sg:person.013361072755.17
115 rdf:rest rdf:nil
116 Nf41d15e2289844819c205576b0af178c schema:name doi
117 schema:value 10.1007/978-3-030-68238-5_31
118 rdf:type schema:PropertyValue
119 anzsrc-for:08 schema:inDefinedTermSet anzsrc-for:
120 schema:name Information and Computing Sciences
121 rdf:type schema:DefinedTerm
122 anzsrc-for:0801 schema:inDefinedTermSet anzsrc-for:
123 schema:name Artificial Intelligence and Image Processing
124 rdf:type schema:DefinedTerm
125 sg:person.013361072755.17 schema:affiliation grid-institutes:grid.507511.7
126 schema:familyName Fritz
127 schema:givenName Mario
128 schema:sameAs https://app.dimensions.ai/discover/publication?and_facet_researcher=ur.013361072755.17
129 rdf:type schema:Person
130 sg:person.013434275371.03 schema:affiliation grid-institutes:grid.419528.3
131 schema:familyName Sattar
132 schema:givenName Hosnieh
133 schema:sameAs https://app.dimensions.ai/discover/publication?and_facet_researcher=ur.013434275371.03
134 rdf:type schema:Person
135 sg:person.015242465107.59 schema:affiliation grid-institutes:grid.419528.3
136 schema:familyName Pons-Moll
137 schema:givenName Gerard
138 schema:sameAs https://app.dimensions.ai/discover/publication?and_facet_researcher=ur.015242465107.59
139 rdf:type schema:Person
140 sg:person.015614373645.33 schema:affiliation grid-institutes:grid.507511.7
141 schema:familyName Krombholz
142 schema:givenName Katharina
143 schema:sameAs https://app.dimensions.ai/discover/publication?and_facet_researcher=ur.015614373645.33
144 rdf:type schema:Person
145 grid-institutes:grid.419528.3 schema:alternateName Max Planck Institute for Informatics, Saarbrücken, Germany
146 schema:name Max Planck Institute for Informatics, Saarbrücken, Germany
147 rdf:type schema:Organization
148 grid-institutes:grid.507511.7 schema:alternateName CISPA Helmholtz Center for Information Security, Saarland Informatics Campus, Saarbrücken, Germany
149 schema:name CISPA Helmholtz Center for Information Security, Saarland Informatics Campus, Saarbrücken, Germany
150 rdf:type schema:Organization
 




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


...