CA‐PMG: Channel attention and progressive multi‐granularity training network for fine‐grained visual classification

Abstract Fine‐grained visual classification is challenging due to the inherently subtle intra‐class object variations. To solve this issue, a novel framework named channel attention and progressive multi‐granularity training network, is proposed. It first exploits meaningful feature maps through the...

Descripción completa

Guardado en:

Detalles Bibliográficos
Autores principales:	Peipei Zhao, Qiguang Miao, Hang Yao, Xiangzeng Liu, Ruyi Liu, Maoguo Gong
Formato:	article
Lenguaje:	EN
Publicado:	Wiley 2021
Materias:	Photography TR1-1050 Computer software QA76.75-76.765
Acceso en línea:	https://doaj.org/article/4cef360c02644bdb9315476c85c48414
Etiquetas:	Agregar Etiqueta Sin Etiquetas, Sea el primero en etiquetar este registro!

id	oai:doaj.org-article:4cef360c02644bdb9315476c85c48414
record_format	dspace
spelling	oai:doaj.org-article:4cef360c02644bdb9315476c85c484142021-11-29T03:38:16ZCA‐PMG: Channel attention and progressive multi‐granularity training network for fine‐grained visual classification1751-96671751-965910.1049/ipr2.12238https://doaj.org/article/4cef360c02644bdb9315476c85c484142021-12-01T00:00:00Zhttps://doi.org/10.1049/ipr2.12238https://doaj.org/toc/1751-9659https://doaj.org/toc/1751-9667Abstract Fine‐grained visual classification is challenging due to the inherently subtle intra‐class object variations. To solve this issue, a novel framework named channel attention and progressive multi‐granularity training network, is proposed. It first exploits meaningful feature maps through the channel attention module and captures multi‐granularity features by the progressive multi‐granularity training module. For each feature map, the channel attention module is proposed to explore channel‐wise correlation. This allows the model to re‐weight the channels of the feature map according to the impact of their semantic information on performance. Furthermore, the progressive multi‐granularity training module is introduced to fuse features cross multi‐granularity. And the fused features pay more attention to the subtle differences between images. The model can be trained efficiently in an end‐to‐end manner without bounding box or part annotations. Finally, comprehensive experiments are conducted to show that the method achieves state‐of‐the‐art performances on the CUB‐200‐2011, Stanford Cars, and FGVC‐Aircraft datasets. Ablation studies demonstrate the effectiveness of each part in our module.Peipei ZhaoQiguang MiaoHang YaoXiangzeng LiuRuyi LiuMaoguo GongWileyarticlePhotographyTR1-1050Computer softwareQA76.75-76.765ENIET Image Processing, Vol 15, Iss 14, Pp 3718-3727 (2021)
institution	DOAJ
collection	DOAJ
language	EN
topic	Photography TR1-1050 Computer software QA76.75-76.765
spellingShingle	Photography TR1-1050 Computer software QA76.75-76.765 Peipei Zhao Qiguang Miao Hang Yao Xiangzeng Liu Ruyi Liu Maoguo Gong CA‐PMG: Channel attention and progressive multi‐granularity training network for fine‐grained visual classification
description	Abstract Fine‐grained visual classification is challenging due to the inherently subtle intra‐class object variations. To solve this issue, a novel framework named channel attention and progressive multi‐granularity training network, is proposed. It first exploits meaningful feature maps through the channel attention module and captures multi‐granularity features by the progressive multi‐granularity training module. For each feature map, the channel attention module is proposed to explore channel‐wise correlation. This allows the model to re‐weight the channels of the feature map according to the impact of their semantic information on performance. Furthermore, the progressive multi‐granularity training module is introduced to fuse features cross multi‐granularity. And the fused features pay more attention to the subtle differences between images. The model can be trained efficiently in an end‐to‐end manner without bounding box or part annotations. Finally, comprehensive experiments are conducted to show that the method achieves state‐of‐the‐art performances on the CUB‐200‐2011, Stanford Cars, and FGVC‐Aircraft datasets. Ablation studies demonstrate the effectiveness of each part in our module.
format	article
author	Peipei Zhao Qiguang Miao Hang Yao Xiangzeng Liu Ruyi Liu Maoguo Gong
author_facet	Peipei Zhao Qiguang Miao Hang Yao Xiangzeng Liu Ruyi Liu Maoguo Gong
author_sort	Peipei Zhao
title	CA‐PMG: Channel attention and progressive multi‐granularity training network for fine‐grained visual classification
title_short	CA‐PMG: Channel attention and progressive multi‐granularity training network for fine‐grained visual classification
title_full	CA‐PMG: Channel attention and progressive multi‐granularity training network for fine‐grained visual classification
title_fullStr	CA‐PMG: Channel attention and progressive multi‐granularity training network for fine‐grained visual classification
title_full_unstemmed	CA‐PMG: Channel attention and progressive multi‐granularity training network for fine‐grained visual classification
title_sort	ca‐pmg: channel attention and progressive multi‐granularity training network for fine‐grained visual classification
publisher	Wiley
publishDate	2021
url	https://doaj.org/article/4cef360c02644bdb9315476c85c48414
work_keys_str_mv	AT peipeizhao capmgchannelattentionandprogressivemultigranularitytrainingnetworkforfinegrainedvisualclassification AT qiguangmiao capmgchannelattentionandprogressivemultigranularitytrainingnetworkforfinegrainedvisualclassification AT hangyao capmgchannelattentionandprogressivemultigranularitytrainingnetworkforfinegrainedvisualclassification AT xiangzengliu capmgchannelattentionandprogressivemultigranularitytrainingnetworkforfinegrainedvisualclassification AT ruyiliu capmgchannelattentionandprogressivemultigranularitytrainingnetworkforfinegrainedvisualclassification AT maoguogong capmgchannelattentionandprogressivemultigranularitytrainingnetworkforfinegrainedvisualclassification
_version_	1718407641444646912

CA‐PMG: Channel attention and progressive multi‐granularity training network for fine‐grained visual classification

Ejemplares similares