CNNs Under Attack: On the Vulnerability of Deep Neural Networks Based Face Recognition to Image Morphing View Full Text


Ontology type: schema:Chapter     


Chapter Info

DATE

2017-07-26

AUTHORS

Lukasz Wandzik , Raul Vicente Garcia , Gerald Kaeding , Xi Chen

ABSTRACT

Facial recognition has become a critical constituent of common automatic border control gates. Despite many advances in recent years, face recognition systems remain susceptible to an ever evolving diversity of spoofing attacks. It has recently been shown that high-quality face morphing or splicing can be employed to deceive facial recognition systems in a border control scenario. Moreover, facial morphs can easily be produced by means of open source software and with minimal technical knowledge. The purpose of this work is to quantify the severeness of the problem using a large dataset of morphed face images. We employ a state-of-the-art face recognition algorithm based on deep convolutional neural networks and measure its performance on a dataset of 7260 high-quality facial morphs with varying blending factor. Using the Inception-ResNet-v1 architecture we train a deep neural model on 4 million images to obtain a validation rate of at false acceptance rate (FAR) on the original, unmodified images. The same model fails to repel of all morphing attacks, accepting both the impostor and the document owner. Based on these results, we discuss the observed weaknesses and possible remedies. More... »

PAGES

121-135

References to SciGraph publications

Book

TITLE

Digital Forensics and Watermarking

ISBN

978-3-319-64184-3
978-3-319-64185-0

Identifiers

URI

http://scigraph.springernature.com/pub.10.1007/978-3-319-64185-0_10

DOI

http://dx.doi.org/10.1007/978-3-319-64185-0_10

DIMENSIONS

https://app.dimensions.ai/details/publication/pub.1090958887


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/2008/0801", 
        "inDefinedTermSet": "http://purl.org/au-research/vocabulary/anzsrc-for/2008/", 
        "name": "Artificial Intelligence and Image Processing", 
        "type": "DefinedTerm"
      }, 
      {
        "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"
      }
    ], 
    "author": [
      {
        "affiliation": {
          "alternateName": "Fraunhofer Institute for Production Systems and Design Technology", 
          "id": "https://www.grid.ac/institutes/grid.469819.b", 
          "name": [
            "Fraunhofer Institute for Production Systems and Design Technology IPK, Pascalstra\u00dfe 8 \u2013 9, 10587, Berlin, Germany"
          ], 
          "type": "Organization"
        }, 
        "familyName": "Wandzik", 
        "givenName": "Lukasz", 
        "id": "sg:person.013570032261.01", 
        "sameAs": [
          "https://app.dimensions.ai/discover/publication?and_facet_researcher=ur.013570032261.01"
        ], 
        "type": "Person"
      }, 
      {
        "affiliation": {
          "alternateName": "Fraunhofer Institute for Production Systems and Design Technology", 
          "id": "https://www.grid.ac/institutes/grid.469819.b", 
          "name": [
            "Fraunhofer Institute for Production Systems and Design Technology IPK, Pascalstra\u00dfe 8 \u2013 9, 10587, Berlin, Germany"
          ], 
          "type": "Organization"
        }, 
        "familyName": "Garcia", 
        "givenName": "Raul Vicente", 
        "id": "sg:person.014365412661.66", 
        "sameAs": [
          "https://app.dimensions.ai/discover/publication?and_facet_researcher=ur.014365412661.66"
        ], 
        "type": "Person"
      }, 
      {
        "affiliation": {
          "alternateName": "Fraunhofer Institute for Production Systems and Design Technology", 
          "id": "https://www.grid.ac/institutes/grid.469819.b", 
          "name": [
            "Fraunhofer Institute for Production Systems and Design Technology IPK, Pascalstra\u00dfe 8 \u2013 9, 10587, Berlin, Germany"
          ], 
          "type": "Organization"
        }, 
        "familyName": "Kaeding", 
        "givenName": "Gerald", 
        "id": "sg:person.015162773261.14", 
        "sameAs": [
          "https://app.dimensions.ai/discover/publication?and_facet_researcher=ur.015162773261.14"
        ], 
        "type": "Person"
      }, 
      {
        "affiliation": {
          "alternateName": "Fraunhofer Institute for Production Systems and Design Technology", 
          "id": "https://www.grid.ac/institutes/grid.469819.b", 
          "name": [
            "Fraunhofer Institute for Production Systems and Design Technology IPK, Pascalstra\u00dfe 8 \u2013 9, 10587, Berlin, Germany"
          ], 
          "type": "Organization"
        }, 
        "familyName": "Chen", 
        "givenName": "Xi", 
        "type": "Person"
      }
    ], 
    "citation": [
      {
        "id": "sg:pub.10.1007/978-3-319-46487-9_6", 
        "sameAs": [
          "https://app.dimensions.ai/details/publication/pub.1006228130", 
          "https://doi.org/10.1007/978-3-319-46487-9_6"
        ], 
        "type": "CreativeWork"
      }, 
      {
        "id": "sg:pub.10.1007/978-3-319-46478-7_31", 
        "sameAs": [
          "https://app.dimensions.ai/details/publication/pub.1026044245", 
          "https://doi.org/10.1007/978-3-319-46478-7_31"
        ], 
        "type": "CreativeWork"
      }, 
      {
        "id": "https://doi.org/10.1109/access.2014.2381273", 
        "sameAs": [
          "https://app.dimensions.ai/details/publication/pub.1061252073"
        ], 
        "type": "CreativeWork"
      }, 
      {
        "id": "https://doi.org/10.1109/lsp.2016.2603342", 
        "sameAs": [
          "https://app.dimensions.ai/details/publication/pub.1061379783"
        ], 
        "type": "CreativeWork"
      }, 
      {
        "id": "https://doi.org/10.1109/msp.2015.2437652", 
        "sameAs": [
          "https://app.dimensions.ai/details/publication/pub.1061424308"
        ], 
        "type": "CreativeWork"
      }, 
      {
        "id": "https://doi.org/10.1145/882262.882269", 
        "sameAs": [
          "https://app.dimensions.ai/details/publication/pub.1063173524"
        ], 
        "type": "CreativeWork"
      }, 
      {
        "id": "https://doi.org/10.1109/btas.2014.6996240", 
        "sameAs": [
          "https://app.dimensions.ai/details/publication/pub.1093614459"
        ], 
        "type": "CreativeWork"
      }, 
      {
        "id": "https://doi.org/10.1109/cvpr.2014.241", 
        "sameAs": [
          "https://app.dimensions.ai/details/publication/pub.1094427070"
        ], 
        "type": "CreativeWork"
      }, 
      {
        "id": "https://doi.org/10.1109/cvpr.2015.7298640", 
        "sameAs": [
          "https://app.dimensions.ai/details/publication/pub.1094486642"
        ], 
        "type": "CreativeWork"
      }, 
      {
        "id": "https://doi.org/10.1109/icip.2010.5650410", 
        "sameAs": [
          "https://app.dimensions.ai/details/publication/pub.1095340004"
        ], 
        "type": "CreativeWork"
      }, 
      {
        "id": "https://doi.org/10.1109/cvprw.2015.7301314", 
        "sameAs": [
          "https://app.dimensions.ai/details/publication/pub.1095402205"
        ], 
        "type": "CreativeWork"
      }, 
      {
        "id": "https://doi.org/10.1109/cvpr.2015.7298682", 
        "sameAs": [
          "https://app.dimensions.ai/details/publication/pub.1095510970"
        ], 
        "type": "CreativeWork"
      }, 
      {
        "id": "https://doi.org/10.5220/0006131100390050", 
        "sameAs": [
          "https://app.dimensions.ai/details/publication/pub.1097043211"
        ], 
        "type": "CreativeWork"
      }, 
      {
        "id": "https://doi.org/10.5244/c.29.41", 
        "sameAs": [
          "https://app.dimensions.ai/details/publication/pub.1099427264"
        ], 
        "type": "CreativeWork"
      }
    ], 
    "datePublished": "2017-07-26", 
    "datePublishedReg": "2017-07-26", 
    "description": "Facial recognition has become a critical constituent of common automatic border control gates. Despite many advances in recent years, face recognition systems remain susceptible to an ever evolving diversity of spoofing attacks. It has recently been shown that high-quality face morphing or splicing can be employed to deceive facial recognition systems in a border control scenario. Moreover, facial morphs can easily be produced by means of open source software and with minimal technical knowledge. The purpose of this work is to quantify the severeness of the problem using a large dataset of morphed face images. We employ a state-of-the-art face recognition algorithm based on deep convolutional neural networks and measure its performance on a dataset of 7260 high-quality facial morphs with varying blending factor. Using the Inception-ResNet-v1 architecture we train a deep neural model on 4 million images to obtain a validation rate of at false acceptance rate (FAR) on the original, unmodified images. The same model fails to repel of all morphing attacks, accepting both the impostor and the document owner. Based on these results, we discuss the observed weaknesses and possible remedies.", 
    "editor": [
      {
        "familyName": "Kraetzer", 
        "givenName": "Christian", 
        "type": "Person"
      }, 
      {
        "familyName": "Shi", 
        "givenName": "Yun-Qing", 
        "type": "Person"
      }, 
      {
        "familyName": "Dittmann", 
        "givenName": "Jana", 
        "type": "Person"
      }, 
      {
        "familyName": "Kim", 
        "givenName": "Hyoung Joong", 
        "type": "Person"
      }
    ], 
    "genre": "chapter", 
    "id": "sg:pub.10.1007/978-3-319-64185-0_10", 
    "inLanguage": [
      "en"
    ], 
    "isAccessibleForFree": false, 
    "isPartOf": {
      "isbn": [
        "978-3-319-64184-3", 
        "978-3-319-64185-0"
      ], 
      "name": "Digital Forensics and Watermarking", 
      "type": "Book"
    }, 
    "name": "CNNs Under Attack: On the Vulnerability of Deep Neural Networks Based Face Recognition to Image Morphing", 
    "pagination": "121-135", 
    "productId": [
      {
        "name": "doi", 
        "type": "PropertyValue", 
        "value": [
          "10.1007/978-3-319-64185-0_10"
        ]
      }, 
      {
        "name": "readcube_id", 
        "type": "PropertyValue", 
        "value": [
          "1ea2780f6511a0871aebc1ebe27830fedbd9fe0c8e29929a1088a10c218c56d8"
        ]
      }, 
      {
        "name": "dimensions_id", 
        "type": "PropertyValue", 
        "value": [
          "pub.1090958887"
        ]
      }
    ], 
    "publisher": {
      "location": "Cham", 
      "name": "Springer International Publishing", 
      "type": "Organisation"
    }, 
    "sameAs": [
      "https://doi.org/10.1007/978-3-319-64185-0_10", 
      "https://app.dimensions.ai/details/publication/pub.1090958887"
    ], 
    "sdDataset": "chapters", 
    "sdDatePublished": "2019-04-16T05:00", 
    "sdLicense": "https://scigraph.springernature.com/explorer/license/", 
    "sdPublisher": {
      "name": "Springer Nature - SN SciGraph project", 
      "type": "Organization"
    }, 
    "sdSource": "s3://com-uberresearch-data-dimensions-target-20181106-alternative/cleanup/v134/2549eaecd7973599484d7c17b260dba0a4ecb94b/merge/v9/a6c9fde33151104705d4d7ff012ea9563521a3ce/jats-lookup/v90/0000000325_0000000325/records_100794_00000000.jsonl", 
    "type": "Chapter", 
    "url": "https://link.springer.com/10.1007%2F978-3-319-64185-0_10"
  }
]
 

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.10.1007/978-3-319-64185-0_10'

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

curl -H 'Accept: application/n-triples' 'https://scigraph.springernature.com/pub.10.1007/978-3-319-64185-0_10'

Turtle is a human-readable linked data format.

curl -H 'Accept: text/turtle' 'https://scigraph.springernature.com/pub.10.1007/978-3-319-64185-0_10'

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

curl -H 'Accept: application/rdf+xml' 'https://scigraph.springernature.com/pub.10.1007/978-3-319-64185-0_10'


 

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

144 TRIPLES      23 PREDICATES      40 URIs      19 LITERALS      8 BLANK NODES

Subject Predicate Object
1 sg:pub.10.1007/978-3-319-64185-0_10 schema:about anzsrc-for:08
2 anzsrc-for:0801
3 schema:author Na67ccc61c25d41a286168efe98bd07c6
4 schema:citation sg:pub.10.1007/978-3-319-46478-7_31
5 sg:pub.10.1007/978-3-319-46487-9_6
6 https://doi.org/10.1109/access.2014.2381273
7 https://doi.org/10.1109/btas.2014.6996240
8 https://doi.org/10.1109/cvpr.2014.241
9 https://doi.org/10.1109/cvpr.2015.7298640
10 https://doi.org/10.1109/cvpr.2015.7298682
11 https://doi.org/10.1109/cvprw.2015.7301314
12 https://doi.org/10.1109/icip.2010.5650410
13 https://doi.org/10.1109/lsp.2016.2603342
14 https://doi.org/10.1109/msp.2015.2437652
15 https://doi.org/10.1145/882262.882269
16 https://doi.org/10.5220/0006131100390050
17 https://doi.org/10.5244/c.29.41
18 schema:datePublished 2017-07-26
19 schema:datePublishedReg 2017-07-26
20 schema:description Facial recognition has become a critical constituent of common automatic border control gates. Despite many advances in recent years, face recognition systems remain susceptible to an ever evolving diversity of spoofing attacks. It has recently been shown that high-quality face morphing or splicing can be employed to deceive facial recognition systems in a border control scenario. Moreover, facial morphs can easily be produced by means of open source software and with minimal technical knowledge. The purpose of this work is to quantify the severeness of the problem using a large dataset of morphed face images. We employ a state-of-the-art face recognition algorithm based on deep convolutional neural networks and measure its performance on a dataset of 7260 high-quality facial morphs with varying blending factor. Using the Inception-ResNet-v1 architecture we train a deep neural model on 4 million images to obtain a validation rate of at false acceptance rate (FAR) on the original, unmodified images. The same model fails to repel of all morphing attacks, accepting both the impostor and the document owner. Based on these results, we discuss the observed weaknesses and possible remedies.
21 schema:editor N1e9167d1f5454458a2b8d0c19cc3d5c9
22 schema:genre chapter
23 schema:inLanguage en
24 schema:isAccessibleForFree false
25 schema:isPartOf N64570928946047a4acbafbaa1474f88e
26 schema:name CNNs Under Attack: On the Vulnerability of Deep Neural Networks Based Face Recognition to Image Morphing
27 schema:pagination 121-135
28 schema:productId N6827189781774a529b10346f4a009adc
29 Nb6db968f71d647f2a28e089a198b164f
30 Nf828533d24b34acbaf09e6d6ae33e742
31 schema:publisher N4415ddb7fead423e95d498ac7235cd05
32 schema:sameAs https://app.dimensions.ai/details/publication/pub.1090958887
33 https://doi.org/10.1007/978-3-319-64185-0_10
34 schema:sdDatePublished 2019-04-16T05:00
35 schema:sdLicense https://scigraph.springernature.com/explorer/license/
36 schema:sdPublisher Nd81d007f06ca45b0bbd821eb608b5e0e
37 schema:url https://link.springer.com/10.1007%2F978-3-319-64185-0_10
38 sgo:license sg:explorer/license/
39 sgo:sdDataset chapters
40 rdf:type schema:Chapter
41 N13d6efea64b64729abedd2c9f0568b9c schema:familyName Kraetzer
42 schema:givenName Christian
43 rdf:type schema:Person
44 N1e9167d1f5454458a2b8d0c19cc3d5c9 rdf:first N13d6efea64b64729abedd2c9f0568b9c
45 rdf:rest N3b1ad6b452984f028fb6455989da4005
46 N20478687633c4e29bf7d64abddf1397f schema:familyName Dittmann
47 schema:givenName Jana
48 rdf:type schema:Person
49 N2103397ae81c415e91e746a02247debc schema:affiliation https://www.grid.ac/institutes/grid.469819.b
50 schema:familyName Chen
51 schema:givenName Xi
52 rdf:type schema:Person
53 N3b1ad6b452984f028fb6455989da4005 rdf:first Ned2d5055309b4e41868c11f5cc893571
54 rdf:rest N5cf44fae82074612a9750454b8b36260
55 N4415ddb7fead423e95d498ac7235cd05 schema:location Cham
56 schema:name Springer International Publishing
57 rdf:type schema:Organisation
58 N5cf44fae82074612a9750454b8b36260 rdf:first N20478687633c4e29bf7d64abddf1397f
59 rdf:rest Nb531017c0f4c4e438da9240dee671af6
60 N64570928946047a4acbafbaa1474f88e schema:isbn 978-3-319-64184-3
61 978-3-319-64185-0
62 schema:name Digital Forensics and Watermarking
63 rdf:type schema:Book
64 N65da5efbbc234fb0b74515eb4bdd6411 rdf:first N2103397ae81c415e91e746a02247debc
65 rdf:rest rdf:nil
66 N6827189781774a529b10346f4a009adc schema:name dimensions_id
67 schema:value pub.1090958887
68 rdf:type schema:PropertyValue
69 N91c3f029ffff483da10ae51f0c4ed41a rdf:first sg:person.014365412661.66
70 rdf:rest Nf20757b1c0c54de1ad0e2a90449b8ee8
71 Na67ccc61c25d41a286168efe98bd07c6 rdf:first sg:person.013570032261.01
72 rdf:rest N91c3f029ffff483da10ae51f0c4ed41a
73 Nb531017c0f4c4e438da9240dee671af6 rdf:first Nba090ed1c3e64fa6997a90a80cbccd3d
74 rdf:rest rdf:nil
75 Nb6db968f71d647f2a28e089a198b164f schema:name doi
76 schema:value 10.1007/978-3-319-64185-0_10
77 rdf:type schema:PropertyValue
78 Nba090ed1c3e64fa6997a90a80cbccd3d schema:familyName Kim
79 schema:givenName Hyoung Joong
80 rdf:type schema:Person
81 Nd81d007f06ca45b0bbd821eb608b5e0e schema:name Springer Nature - SN SciGraph project
82 rdf:type schema:Organization
83 Ned2d5055309b4e41868c11f5cc893571 schema:familyName Shi
84 schema:givenName Yun-Qing
85 rdf:type schema:Person
86 Nf20757b1c0c54de1ad0e2a90449b8ee8 rdf:first sg:person.015162773261.14
87 rdf:rest N65da5efbbc234fb0b74515eb4bdd6411
88 Nf828533d24b34acbaf09e6d6ae33e742 schema:name readcube_id
89 schema:value 1ea2780f6511a0871aebc1ebe27830fedbd9fe0c8e29929a1088a10c218c56d8
90 rdf:type schema:PropertyValue
91 anzsrc-for:08 schema:inDefinedTermSet anzsrc-for:
92 schema:name Information and Computing Sciences
93 rdf:type schema:DefinedTerm
94 anzsrc-for:0801 schema:inDefinedTermSet anzsrc-for:
95 schema:name Artificial Intelligence and Image Processing
96 rdf:type schema:DefinedTerm
97 sg:person.013570032261.01 schema:affiliation https://www.grid.ac/institutes/grid.469819.b
98 schema:familyName Wandzik
99 schema:givenName Lukasz
100 schema:sameAs https://app.dimensions.ai/discover/publication?and_facet_researcher=ur.013570032261.01
101 rdf:type schema:Person
102 sg:person.014365412661.66 schema:affiliation https://www.grid.ac/institutes/grid.469819.b
103 schema:familyName Garcia
104 schema:givenName Raul Vicente
105 schema:sameAs https://app.dimensions.ai/discover/publication?and_facet_researcher=ur.014365412661.66
106 rdf:type schema:Person
107 sg:person.015162773261.14 schema:affiliation https://www.grid.ac/institutes/grid.469819.b
108 schema:familyName Kaeding
109 schema:givenName Gerald
110 schema:sameAs https://app.dimensions.ai/discover/publication?and_facet_researcher=ur.015162773261.14
111 rdf:type schema:Person
112 sg:pub.10.1007/978-3-319-46478-7_31 schema:sameAs https://app.dimensions.ai/details/publication/pub.1026044245
113 https://doi.org/10.1007/978-3-319-46478-7_31
114 rdf:type schema:CreativeWork
115 sg:pub.10.1007/978-3-319-46487-9_6 schema:sameAs https://app.dimensions.ai/details/publication/pub.1006228130
116 https://doi.org/10.1007/978-3-319-46487-9_6
117 rdf:type schema:CreativeWork
118 https://doi.org/10.1109/access.2014.2381273 schema:sameAs https://app.dimensions.ai/details/publication/pub.1061252073
119 rdf:type schema:CreativeWork
120 https://doi.org/10.1109/btas.2014.6996240 schema:sameAs https://app.dimensions.ai/details/publication/pub.1093614459
121 rdf:type schema:CreativeWork
122 https://doi.org/10.1109/cvpr.2014.241 schema:sameAs https://app.dimensions.ai/details/publication/pub.1094427070
123 rdf:type schema:CreativeWork
124 https://doi.org/10.1109/cvpr.2015.7298640 schema:sameAs https://app.dimensions.ai/details/publication/pub.1094486642
125 rdf:type schema:CreativeWork
126 https://doi.org/10.1109/cvpr.2015.7298682 schema:sameAs https://app.dimensions.ai/details/publication/pub.1095510970
127 rdf:type schema:CreativeWork
128 https://doi.org/10.1109/cvprw.2015.7301314 schema:sameAs https://app.dimensions.ai/details/publication/pub.1095402205
129 rdf:type schema:CreativeWork
130 https://doi.org/10.1109/icip.2010.5650410 schema:sameAs https://app.dimensions.ai/details/publication/pub.1095340004
131 rdf:type schema:CreativeWork
132 https://doi.org/10.1109/lsp.2016.2603342 schema:sameAs https://app.dimensions.ai/details/publication/pub.1061379783
133 rdf:type schema:CreativeWork
134 https://doi.org/10.1109/msp.2015.2437652 schema:sameAs https://app.dimensions.ai/details/publication/pub.1061424308
135 rdf:type schema:CreativeWork
136 https://doi.org/10.1145/882262.882269 schema:sameAs https://app.dimensions.ai/details/publication/pub.1063173524
137 rdf:type schema:CreativeWork
138 https://doi.org/10.5220/0006131100390050 schema:sameAs https://app.dimensions.ai/details/publication/pub.1097043211
139 rdf:type schema:CreativeWork
140 https://doi.org/10.5244/c.29.41 schema:sameAs https://app.dimensions.ai/details/publication/pub.1099427264
141 rdf:type schema:CreativeWork
142 https://www.grid.ac/institutes/grid.469819.b schema:alternateName Fraunhofer Institute for Production Systems and Design Technology
143 schema:name Fraunhofer Institute for Production Systems and Design Technology IPK, Pascalstraße 8 – 9, 10587, Berlin, Germany
144 rdf:type schema:Organization
 




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


...