Fine-grained feature extraction method, fine-grained image recognition method and related devices

The invention discloses a fine-grained feature extraction method, a fine-grained image recognition method and a related device, a graph neural network is added into a framework for decomposing bilinear pooling, similar features in a spatial domain can be fused, various visual features are fully util...

Full description

Saved in:
Bibliographic Details
Main Authors WANG DONG, REN JUNHONG, SUN MAOMAO, RUI TING, DU YINAN, ZENG YONGHUA, ZHENG NAN, DU XIAOMING, FANG HUSHENG, YANG CHENGSONG, LIU XUN
Format Patent
LanguageChinese
English
Published 22.03.2024
Subjects
CALCULATING
COMPUTER SYSTEMS BASED ON SPECIFIC COMPUTATIONAL MODELS
COMPUTING
COUNTING
PHYSICS
Online AccessGet full text

Cover

More Information
Summary:The invention discloses a fine-grained feature extraction method, a fine-grained image recognition method and a related device, a graph neural network is added into a framework for decomposing bilinear pooling, similar features in a spatial domain can be fused, various visual features are fully utilized, and a visual burst phenomenon of bilinear pooling is relieved. 本发明公开了一种细粒度特征提取方法、细粒度图像识别方法及相关装置，本发明在分解双线性池化的框架中加入图神经网络，可融合空域上的相似特征，实现对多种视觉特征的充分利用，缓解双线性池化的视觉突发现象。
Bibliography:Application Number: CN202311767985