Understanding fake faces

Ryota Natsume, Kazuki Inoue, Yoshihiro Fukuhara, Shintaro Yamamoto, Shigeo Morishima, Hirokatsu Kataoka

    研究成果: Conference contribution

    抄録

    Face recognition research is one of the most active topics in computer vision (CV), and deep neural networks (DNN) are now filling the gap between human-level and computer-driven performance levels in face verification algorithms. However, although the performance gap appears to be narrowing in terms of accuracy-based expectations, a curious question has arisen; specifically, Face understanding of AI is really close to that of human? In the present study, in an effort to confirm the brain-driven concept, we conduct image-based detection, classification, and generation using an in-house created fake face database. This database has two configurations: (i) false positive face detections produced using both the Viola Jones (VJ) method and convolutional neural networks (CNN), and (ii) simulacra that have fundamental characteristics that resemble faces but are completely artificial. The results show a level of suggestive knowledge that indicates the continuing existence of a gap between the capabilities of recent vision-based face recognition algorithms and human-level performance. On a positive note, however, we have obtained knowledge that will advance the progress of face-understanding models.

    元の言語English
    ホスト出版物のタイトルComputer Vision – ECCV 2018 Workshops, Proceedings
    編集者Laura Leal-Taixé, Stefan Roth
    出版者Springer-Verlag
    ページ566-576
    ページ数11
    ISBN(印刷物)9783030110147
    DOI
    出版物ステータスPublished - 2019 1 1
    イベント15th European Conference on Computer Vision, ECCV 2018 - Munich, Germany
    継続期間: 2018 9 82018 9 14

    出版物シリーズ

    名前Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)
    11131 LNCS
    ISSN(印刷物)0302-9743
    ISSN(電子版)1611-3349

    Conference

    Conference15th European Conference on Computer Vision, ECCV 2018
    Germany
    Munich
    期間18/9/818/9/14

    Fingerprint

    Face recognition
    Face
    Face Recognition
    Computer vision
    Neural Networks
    Brain
    Face Detection
    Neural networks
    Recognition Algorithm
    False Positive
    Computer Vision
    Configuration
    Human
    Knowledge

    Keywords

      ASJC Scopus subject areas

      • Theoretical Computer Science
      • Computer Science(all)

      これを引用

      Natsume, R., Inoue, K., Fukuhara, Y., Yamamoto, S., Morishima, S., & Kataoka, H. (2019). Understanding fake faces. : L. Leal-Taixé, & S. Roth (版), Computer Vision – ECCV 2018 Workshops, Proceedings (pp. 566-576). (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics); 巻数 11131 LNCS). Springer-Verlag. https://doi.org/10.1007/978-3-030-11015-4_42

      Understanding fake faces. / Natsume, Ryota; Inoue, Kazuki; Fukuhara, Yoshihiro; Yamamoto, Shintaro; Morishima, Shigeo; Kataoka, Hirokatsu.

      Computer Vision – ECCV 2018 Workshops, Proceedings. 版 / Laura Leal-Taixé; Stefan Roth. Springer-Verlag, 2019. p. 566-576 (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics); 巻 11131 LNCS).

      研究成果: Conference contribution

      Natsume, R, Inoue, K, Fukuhara, Y, Yamamoto, S, Morishima, S & Kataoka, H 2019, Understanding fake faces. : L Leal-Taixé & S Roth (版), Computer Vision – ECCV 2018 Workshops, Proceedings. Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics), 巻. 11131 LNCS, Springer-Verlag, pp. 566-576, 15th European Conference on Computer Vision, ECCV 2018, Munich, Germany, 18/9/8. https://doi.org/10.1007/978-3-030-11015-4_42
      Natsume R, Inoue K, Fukuhara Y, Yamamoto S, Morishima S, Kataoka H. Understanding fake faces. : Leal-Taixé L, Roth S, 編集者, Computer Vision – ECCV 2018 Workshops, Proceedings. Springer-Verlag. 2019. p. 566-576. (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)). https://doi.org/10.1007/978-3-030-11015-4_42
      Natsume, Ryota ; Inoue, Kazuki ; Fukuhara, Yoshihiro ; Yamamoto, Shintaro ; Morishima, Shigeo ; Kataoka, Hirokatsu. / Understanding fake faces. Computer Vision – ECCV 2018 Workshops, Proceedings. 編集者 / Laura Leal-Taixé ; Stefan Roth. Springer-Verlag, 2019. pp. 566-576 (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)).
      @inproceedings{6b431d915bd74b2db916d7e773727f94,
      title = "Understanding fake faces",
      abstract = "Face recognition research is one of the most active topics in computer vision (CV), and deep neural networks (DNN) are now filling the gap between human-level and computer-driven performance levels in face verification algorithms. However, although the performance gap appears to be narrowing in terms of accuracy-based expectations, a curious question has arisen; specifically, Face understanding of AI is really close to that of human? In the present study, in an effort to confirm the brain-driven concept, we conduct image-based detection, classification, and generation using an in-house created fake face database. This database has two configurations: (i) false positive face detections produced using both the Viola Jones (VJ) method and convolutional neural networks (CNN), and (ii) simulacra that have fundamental characteristics that resemble faces but are completely artificial. The results show a level of suggestive knowledge that indicates the continuing existence of a gap between the capabilities of recent vision-based face recognition algorithms and human-level performance. On a positive note, however, we have obtained knowledge that will advance the progress of face-understanding models.",
      keywords = "Face recognition, False positives, Simulacra",
      author = "Ryota Natsume and Kazuki Inoue and Yoshihiro Fukuhara and Shintaro Yamamoto and Shigeo Morishima and Hirokatsu Kataoka",
      year = "2019",
      month = "1",
      day = "1",
      doi = "10.1007/978-3-030-11015-4_42",
      language = "English",
      isbn = "9783030110147",
      series = "Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)",
      publisher = "Springer-Verlag",
      pages = "566--576",
      editor = "Laura Leal-Taix{\'e} and Stefan Roth",
      booktitle = "Computer Vision – ECCV 2018 Workshops, Proceedings",

      }

      TY - GEN

      T1 - Understanding fake faces

      AU - Natsume, Ryota

      AU - Inoue, Kazuki

      AU - Fukuhara, Yoshihiro

      AU - Yamamoto, Shintaro

      AU - Morishima, Shigeo

      AU - Kataoka, Hirokatsu

      PY - 2019/1/1

      Y1 - 2019/1/1

      N2 - Face recognition research is one of the most active topics in computer vision (CV), and deep neural networks (DNN) are now filling the gap between human-level and computer-driven performance levels in face verification algorithms. However, although the performance gap appears to be narrowing in terms of accuracy-based expectations, a curious question has arisen; specifically, Face understanding of AI is really close to that of human? In the present study, in an effort to confirm the brain-driven concept, we conduct image-based detection, classification, and generation using an in-house created fake face database. This database has two configurations: (i) false positive face detections produced using both the Viola Jones (VJ) method and convolutional neural networks (CNN), and (ii) simulacra that have fundamental characteristics that resemble faces but are completely artificial. The results show a level of suggestive knowledge that indicates the continuing existence of a gap between the capabilities of recent vision-based face recognition algorithms and human-level performance. On a positive note, however, we have obtained knowledge that will advance the progress of face-understanding models.

      AB - Face recognition research is one of the most active topics in computer vision (CV), and deep neural networks (DNN) are now filling the gap between human-level and computer-driven performance levels in face verification algorithms. However, although the performance gap appears to be narrowing in terms of accuracy-based expectations, a curious question has arisen; specifically, Face understanding of AI is really close to that of human? In the present study, in an effort to confirm the brain-driven concept, we conduct image-based detection, classification, and generation using an in-house created fake face database. This database has two configurations: (i) false positive face detections produced using both the Viola Jones (VJ) method and convolutional neural networks (CNN), and (ii) simulacra that have fundamental characteristics that resemble faces but are completely artificial. The results show a level of suggestive knowledge that indicates the continuing existence of a gap between the capabilities of recent vision-based face recognition algorithms and human-level performance. On a positive note, however, we have obtained knowledge that will advance the progress of face-understanding models.

      KW - Face recognition

      KW - False positives

      KW - Simulacra

      UR - http://www.scopus.com/inward/record.url?scp=85061694140&partnerID=8YFLogxK

      UR - http://www.scopus.com/inward/citedby.url?scp=85061694140&partnerID=8YFLogxK

      U2 - 10.1007/978-3-030-11015-4_42

      DO - 10.1007/978-3-030-11015-4_42

      M3 - Conference contribution

      SN - 9783030110147

      T3 - Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)

      SP - 566

      EP - 576

      BT - Computer Vision – ECCV 2018 Workshops, Proceedings

      A2 - Leal-Taixé, Laura

      A2 - Roth, Stefan

      PB - Springer-Verlag

      ER -