• 研究課題をさがす
  • 研究者をさがす
  • KAKENの使い方
  1. 課題ページに戻る

2021 年度 実施状況報告書

Prevention from Automated Analysis Services with Object-Level Adversarial Examples

研究課題

研究課題/領域番号 21K18023
研究機関国立情報学研究所

研究代表者

レ チュンギア  国立情報学研究所, 情報社会相関研究系, 特任研究員 (00884404)

研究期間 (年度) 2021-04-01 – 2023-03-31
キーワードAdversarial examples / Privacy protection
研究実績の概要

We proposed an adversarial example based method for attacking human instance segmentation networks. We developed a novel method to automatically identify attackable regions in the target image to minimize the effect on image quality. The fashion-guided synthesized adversarial textures are inconspicuous and appear natural to the human eye. The effectiveness of the proposed method is enhanced by robustness training and by jointly attacking multiple components of the
target network. This work was published at Workshop on Media Forensics, CVPR 2021.

We analyzed class-aware transferability of adversarial examples to show the strong connection between non-targeted transferability of adversarial examples and same mistakes. Adversarial examples can have non-robust features that correlate with a certain class to which models can be misled. However, different mistakes occur between very similar models regardless of the perturbation size, which raises the question how adversarial examples cause different mistakes. We also demonstrated that non-robust features can comprehensively explain the difference between a different mistake and a same mistake by extending the framework of Ilyas et al. They showed that adversarial examples can have non-robust features that are predictive but human-imperceptible, which can cause a same mistake. In contrast, we showed that when the manipulated non-robust features in an adversarial examples are differently used by multiple models, those models may classify the adversarial examples differently. This work is submitted to ACM MM 2022 and under review.

現在までの達成度 (区分)
現在までの達成度 (区分)

3: やや遅れている

理由

We published a paper about adversarial attack to person segmentation at Workshop on Media Forensics, CVPR 2021. Another paper about adversarial analysis was submitted to ACM MM 2022 and under review. We are currently developing a new adversarial attack method targeting scene recognition systems.

今後の研究の推進方策

We are developing a new adversarial attack method targeting scene recognition systems. We expect that the proposed method can be transferable and invisible. We will also develop new attack methods targeting vision-language systems in the next fiscal year.

次年度使用額が生じた理由

Because of COVID pandemic, conferences have been held as virtual meetings. Therefore, I would like to transfer the budget to the next fiscal year. The next fiscal year's budget and transferred one will be mainly used for equipment, experiments, and conference fees.

  • 研究成果

    (5件)

すべて 2022 2021

すべて 雑誌論文 (5件) (うち国際共著 5件、 査読あり 5件)

  • [雑誌論文] Trung-Nghia Le, Tam V. Nguyen, Minh-Triet Tran2022

    • 著者名/発表者名
      Contextual Guided Segmentation Framework for Semi-supervised Video Instance Segmentation
    • 雑誌名

      Machine Vision and Applications

      巻: 33 ページ: 1-19

    • DOI

      10.1007/s00138-022-01278-x

    • 査読あり / 国際共著
  • [雑誌論文] Fashion-Guided Adversarial Attack on Person Segmentation2021

    • 著者名/発表者名
      Marc Treu, Trung-Nghia Le, Huy H. Nguyen, Junichi Yamagishi, Isao Echizen
    • 雑誌名

      IEEE/CVF Conference on Computer Vision and Pattern Recognition Workshops

      巻: 1 ページ: 943-952

    • DOI

      10.1109/CVPRW53098.2021.00105

    • 査読あり / 国際共著
  • [雑誌論文] OpenForensics: Large-Scale Challenging Dataset For Multi-Face Forgery Detection And Segmentation In-The-Wild2021

    • 著者名/発表者名
      Trung-Nghia Le, Huy H. Nguyen, Junichi Yamagishi, Isao Echizen
    • 雑誌名

      IEEE/CVF International Conference on Computer Vision

      巻: 1 ページ: 10117-10127

    • DOI

      10.1109/ICCV48922.2021.00996

    • 査読あり / 国際共著
  • [雑誌論文] Khanh-Duy Nguyen, Huy H. Nguyen, Trung-Nghia Le, Junichi Yamagishi, Isao Echizen2021

    • 著者名/発表者名
      Effectiveness of Detection-based and Regression-based Approaches for Estimating Mask-Wearing Ratio
    • 雑誌名

      IEEE International Conference on Automatic Face and Gesture Recognition Workshops

      巻: 1 ページ: 1-8

    • DOI

      10.1109/FG52635.2021.9667046

    • 査読あり / 国際共著
  • [雑誌論文] Trung-Nghia Le, Yubo Cao, Tan-Cong Nguyen, Minh-Quan Le, Khanh-Duy Nguyen, Thanh-Toan Do, Minh-Triet Tran, Tam V. Nguyen2021

    • 著者名/発表者名
      Camouflaged Instance Segmentation In-The-Wild: Dataset, Method, and Benchmark Suite
    • 雑誌名

      IEEE Transactions on Image Processing

      巻: 31 ページ: 287-300

    • DOI

      10.1109/TIP.2021.3130490

    • 査読あり / 国際共著

URL: 

公開日: 2022-12-28  

サービス概要 検索マニュアル よくある質問 お知らせ 利用規程 科研費による研究の帰属

Powered by NII kakenhi