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
Les régions discriminantes locales jouent un rôle important dans les tâches d'analyse d'images à granularité fine. Comment localiser les régions discriminantes locales avec uniquement une étiquette de catégorie et apprendre la représentation discriminante de ces régions ont été des points chauds. Dans notre travail, nous proposons la méthode de recherche de régions discriminantes (SDR) et d'apprentissage de régions discriminantes (LDR) pour rechercher et apprendre des régions discriminantes locales dans les images. La méthode SDR adopte un mécanisme d'attention pour rechercher de manière itérative des régions à réponse élevée dans les images et l'utilise comme indice pour localiser les régions discriminantes locales. De plus, la méthode LDR est proposée pour apprendre la représentation compacte au sein d’une catégorie et clairsemée entre les catégories à partir de l’image brute et des images locales. Les résultats expérimentaux montrent que notre approche proposée atteint d'excellentes performances dans les tâches de récupération d'images fines et de classification, ce qui démontre son efficacité.
Kangbo SUN
Shanghai Jiao Tong University,Shanghai Frontier Science Research Center for Gravitational Wave Detection
Jie ZHU
Shanghai Jiao Tong University,Shanghai Frontier Science Research Center for Gravitational Wave Detection
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.
Copier
Kangbo SUN, Jie ZHU, "Searching and Learning Discriminative Regions for Fine-Grained Image Retrieval and Classification" in IEICE TRANSACTIONS on Information,
vol. E105-D, no. 1, pp. 141-149, January 2022, doi: 10.1587/transinf.2021EDP7094.
Abstract: Local discriminative regions play important roles in fine-grained image analysis tasks. How to locate local discriminative regions with only category label and learn discriminative representation from these regions have been hot spots. In our work, we propose Searching Discriminative Regions (SDR) and Learning Discriminative Regions (LDR) method to search and learn local discriminative regions in images. The SDR method adopts attention mechanism to iteratively search for high-response regions in images, and uses this as a clue to locate local discriminative regions. Moreover, the LDR method is proposed to learn compact within category and sparse between categories representation from the raw image and local images. Experimental results show that our proposed approach achieves excellent performance in both fine-grained image retrieval and classification tasks, which demonstrates its effectiveness.
URL: https://global.ieice.org/en_transactions/information/10.1587/transinf.2021EDP7094/_p
Copier
@ARTICLE{e105-d_1_141,
author={Kangbo SUN, Jie ZHU, },
journal={IEICE TRANSACTIONS on Information},
title={Searching and Learning Discriminative Regions for Fine-Grained Image Retrieval and Classification},
year={2022},
volume={E105-D},
number={1},
pages={141-149},
abstract={Local discriminative regions play important roles in fine-grained image analysis tasks. How to locate local discriminative regions with only category label and learn discriminative representation from these regions have been hot spots. In our work, we propose Searching Discriminative Regions (SDR) and Learning Discriminative Regions (LDR) method to search and learn local discriminative regions in images. The SDR method adopts attention mechanism to iteratively search for high-response regions in images, and uses this as a clue to locate local discriminative regions. Moreover, the LDR method is proposed to learn compact within category and sparse between categories representation from the raw image and local images. Experimental results show that our proposed approach achieves excellent performance in both fine-grained image retrieval and classification tasks, which demonstrates its effectiveness.},
keywords={},
doi={10.1587/transinf.2021EDP7094},
ISSN={1745-1361},
month={January},}
Copier
TY - JOUR
TI - Searching and Learning Discriminative Regions for Fine-Grained Image Retrieval and Classification
T2 - IEICE TRANSACTIONS on Information
SP - 141
EP - 149
AU - Kangbo SUN
AU - Jie ZHU
PY - 2022
DO - 10.1587/transinf.2021EDP7094
JO - IEICE TRANSACTIONS on Information
SN - 1745-1361
VL - E105-D
IS - 1
JA - IEICE TRANSACTIONS on Information
Y1 - January 2022
AB - Local discriminative regions play important roles in fine-grained image analysis tasks. How to locate local discriminative regions with only category label and learn discriminative representation from these regions have been hot spots. In our work, we propose Searching Discriminative Regions (SDR) and Learning Discriminative Regions (LDR) method to search and learn local discriminative regions in images. The SDR method adopts attention mechanism to iteratively search for high-response regions in images, and uses this as a clue to locate local discriminative regions. Moreover, the LDR method is proposed to learn compact within category and sparse between categories representation from the raw image and local images. Experimental results show that our proposed approach achieves excellent performance in both fine-grained image retrieval and classification tasks, which demonstrates its effectiveness.
ER -