CASIA OpenIR  > 多模态人工智能系统全国重点实验室  > 机器人视觉
Hierarchical Attention Network for Open-Set Fine-Grained Recognition
Jiayin, Sun1,2,3; Hong, Wang3,4; Qiulei, Dong1,2,3
Source PublicationIEEE Transactions on Circuits and Systems for Video Technology
2023-10
Pages1-14
Abstract

Triggered by the success of transformers in various visual tasks, the spatial self-attention mechanism has recently attracted more and more attention in the computer vision community. However, we empirically found that a typical vision transformer with the spatial self-attention mechanism could not learn accurate attention maps for distinguishing different categories of fine-grained images. To address this problem, motivated by the temporal attention mechanism in brains, we propose a hierarchical attention network for learning fine-grained feature representations, called HAN, where the features learnt by implementing a sequence of spatial self-attention operations corresponding to multiple moments are aggregated progressively. The proposed HAN consists of four modules: a self-attention backbone module for learning a sequence of features with selfattention operations, a spatial feature self-organizing module for facilitating the model training, a hierarchical aggregation module for aggregating the re-organized features via a Long Short-Term Memory network, and a context-aware module that is implemented as the forget block of the hierarchical aggregation module for preserving/forgetting the long-term memory by utilizing contextual information. Then, we propose a HAN-based method for open-set fine-grained recognition by integrating the proposed HAN network with a linear classifier, called HAN-OSFGR. Extensive experimental results on 3 fine-grained datasets and 2 coarse-grained datasets demonstrate that the proposed HAN-OSFGR outperforms 9 state-of-the-art open-set recognition methods significantly in most cases.

Indexed BySCI
Language英语
IS Representative Paper
Sub direction classification目标检测、跟踪与识别
planning direction of the national heavy laboratory多尺度信息处理
Paper associated data
Document Type期刊论文
Identifierhttp://ir.ia.ac.cn/handle/173211/56588
Collection多模态人工智能系统全国重点实验室_机器人视觉
Corresponding AuthorQiulei, Dong
Affiliation1.Chinese Acad Sci, Inst Automat, Natl Lab Pattern Recognit, Beijing 100190, Peoples R China
2.Chinese Acad Sci, Ctr Excellence Brain Sci & Intelligence Technol, Beijing 100190, Peoples R China
3.Univ Chinese Acad Sci, Sch Artificial Intelligence, Beijing 100049, Peoples R China
4.Univ Chinese Acad Sci, Coll Life Sci, Beijing 100049, Peoples R China
First Author AffilicationChinese Acad Sci, Inst Automat, Natl Lab Pattern Recognit, Beijing 100190, Peoples R China
Corresponding Author AffilicationChinese Acad Sci, Inst Automat, Natl Lab Pattern Recognit, Beijing 100190, Peoples R China
Recommended Citation
GB/T 7714
Jiayin, Sun,Hong, Wang,Qiulei, Dong. Hierarchical Attention Network for Open-Set Fine-Grained Recognition[J]. IEEE Transactions on Circuits and Systems for Video Technology,2023:1-14.
APA Jiayin, Sun,Hong, Wang,&Qiulei, Dong.(2023).Hierarchical Attention Network for Open-Set Fine-Grained Recognition.IEEE Transactions on Circuits and Systems for Video Technology,1-14.
MLA Jiayin, Sun,et al."Hierarchical Attention Network for Open-Set Fine-Grained Recognition".IEEE Transactions on Circuits and Systems for Video Technology (2023):1-14.
Files in This Item:
File Name/Size DocType Version Access License
Hierarchical_Attenti(2596KB)期刊论文作者接受稿开放获取CC BY-NC-SAView
Related Services
Recommend this item
Bookmark
Usage statistics
Export to Endnote
Google Scholar
Similar articles in Google Scholar
[Jiayin, Sun]'s Articles
[Hong, Wang]'s Articles
[Qiulei, Dong]'s Articles
Baidu academic
Similar articles in Baidu academic
[Jiayin, Sun]'s Articles
[Hong, Wang]'s Articles
[Qiulei, Dong]'s Articles
Bing Scholar
Similar articles in Bing Scholar
[Jiayin, Sun]'s Articles
[Hong, Wang]'s Articles
[Qiulei, Dong]'s Articles
Terms of Use
No data!
Social Bookmark/Share
File name: Hierarchical_Attention_Network_for_Open-Set_Fine-Grained_Image_Recognition.pdf
Format: Adobe PDF
All comments (0)
No comment.
 

Items in the repository are protected by copyright, with all rights reserved, unless otherwise indicated.