The original paper is in English. Non-English content has been machine-translated and may contain typographical errors or mistranslations. ex. Some numerals are expressed as "XNUMX".
Copyrights notice
The original paper is in English. Non-English content has been machine-translated and may contain typographical errors or mistranslations. Copyrights notice
Baru-baru ini, ciri tempatan yang dikira menggunakan rangkaian neural convolutional (CNN) menunjukkan prestasi yang baik untuk mendapatkan semula imej. Ciri konvolusi tempatan yang diperolehi oleh CNN (ciri LC) direka bentuk untuk menjadi invarian terjemahan, walau bagaimanapun, ia sememangnya sensitif terhadap gangguan putaran. Ini membawa kepada kesilapan penghakiman dalam tugas mendapatkan semula. Dalam kerja ini, objektif kami adalah untuk meningkatkan keteguhan ciri LC terhadap putaran imej. Untuk melakukan ini, kami menjalankan penilaian percubaan menyeluruh terhadap tiga strategi anti-putaran calon (pembesaran data dalam model, penambahan ciri dalam model dan penambahan ciri pasca model), ke atas dua jenis serangan putaran (serangan set data dan serangan pertanyaan. ). Dalam prosedur latihan, kami melaksanakan protokol penambahan data dan kaedah penambahan rangkaian. Dalam prosedur ujian, kami membangunkan kaedah pengekstrakan ciri konvolusi berubah tempatan (LTC) dan menilainya melalui konfigurasi rangkaian yang berbeza. Kami menamatkan satu siri amalan baik dengan sokongan kuantitatif yang mantap, yang membawa kepada strategi terbaik untuk pengiraan ciri LC dengan invarian putaran tinggi dalam pengambilan imej.
Longjiao ZHAO
Nagoya University
Yu WANG
Ritsumeikan University
Jien KATO
Ritsumeikan University
The copyright of the original papers published on this site belongs to IEICE. Unauthorized use of the original or translated papers is prohibited. See IEICE Provisions on Copyright for details.
Salinan
Longjiao ZHAO, Yu WANG, Jien KATO, "Rethinking the Rotation Invariance of Local Convolutional Features for Content-Based Image Retrieval" in IEICE TRANSACTIONS on Information,
vol. E104-D, no. 1, pp. 174-182, January 2021, doi: 10.1587/transinf.2020EDP7017.
Abstract: Recently, local features computed using convolutional neural networks (CNNs) show good performance to image retrieval. The local convolutional features obtained by the CNNs (LC features) are designed to be translation invariant, however, they are inherently sensitive to rotation perturbations. This leads to miss-judgements in retrieval tasks. In this work, our objective is to enhance the robustness of LC features against image rotation. To do this, we conduct a thorough experimental evaluation of three candidate anti-rotation strategies (in-model data augmentation, in-model feature augmentation, and post-model feature augmentation), over two kinds of rotation attack (dataset attack and query attack). In the training procedure, we implement a data augmentation protocol and network augmentation method. In the test procedure, we develop a local transformed convolutional (LTC) feature extraction method, and evaluate it over different network configurations. We end up a series of good practices with steady quantitative supports, which lead to the best strategy for computing LC features with high rotation invariance in image retrieval.
URL: https://global.ieice.org/en_transactions/information/10.1587/transinf.2020EDP7017/_p
Salinan
@ARTICLE{e104-d_1_174,
author={Longjiao ZHAO, Yu WANG, Jien KATO, },
journal={IEICE TRANSACTIONS on Information},
title={Rethinking the Rotation Invariance of Local Convolutional Features for Content-Based Image Retrieval},
year={2021},
volume={E104-D},
number={1},
pages={174-182},
abstract={Recently, local features computed using convolutional neural networks (CNNs) show good performance to image retrieval. The local convolutional features obtained by the CNNs (LC features) are designed to be translation invariant, however, they are inherently sensitive to rotation perturbations. This leads to miss-judgements in retrieval tasks. In this work, our objective is to enhance the robustness of LC features against image rotation. To do this, we conduct a thorough experimental evaluation of three candidate anti-rotation strategies (in-model data augmentation, in-model feature augmentation, and post-model feature augmentation), over two kinds of rotation attack (dataset attack and query attack). In the training procedure, we implement a data augmentation protocol and network augmentation method. In the test procedure, we develop a local transformed convolutional (LTC) feature extraction method, and evaluate it over different network configurations. We end up a series of good practices with steady quantitative supports, which lead to the best strategy for computing LC features with high rotation invariance in image retrieval.},
keywords={},
doi={10.1587/transinf.2020EDP7017},
ISSN={1745-1361},
month={January},}
Salinan
TY - JOUR
TI - Rethinking the Rotation Invariance of Local Convolutional Features for Content-Based Image Retrieval
T2 - IEICE TRANSACTIONS on Information
SP - 174
EP - 182
AU - Longjiao ZHAO
AU - Yu WANG
AU - Jien KATO
PY - 2021
DO - 10.1587/transinf.2020EDP7017
JO - IEICE TRANSACTIONS on Information
SN - 1745-1361
VL - E104-D
IS - 1
JA - IEICE TRANSACTIONS on Information
Y1 - January 2021
AB - Recently, local features computed using convolutional neural networks (CNNs) show good performance to image retrieval. The local convolutional features obtained by the CNNs (LC features) are designed to be translation invariant, however, they are inherently sensitive to rotation perturbations. This leads to miss-judgements in retrieval tasks. In this work, our objective is to enhance the robustness of LC features against image rotation. To do this, we conduct a thorough experimental evaluation of three candidate anti-rotation strategies (in-model data augmentation, in-model feature augmentation, and post-model feature augmentation), over two kinds of rotation attack (dataset attack and query attack). In the training procedure, we implement a data augmentation protocol and network augmentation method. In the test procedure, we develop a local transformed convolutional (LTC) feature extraction method, and evaluate it over different network configurations. We end up a series of good practices with steady quantitative supports, which lead to the best strategy for computing LC features with high rotation invariance in image retrieval.
ER -