Meta label associated loss for fine-grained visual recognition

扫码查看

原文链接

NETL
NSTL
万方数据

外文摘要：Recently,intensive attempts have been made to design robust models for fine-grained visual recognition,most notably are the impressive gains for training with noisy labels by incorporating a reweighting strategy into a meta-learning framework.However,it is limited to up or downweighting the contribution of an instance for label reweighting approaches in the learning process.To solve this issue,a novel noise-tolerant method with auxiliary web data is proposed.Specifically,first,the associations made from embeddings of well-labeled data with those of web data and back at the same class are measured.Next,its association probability is employed as a weighting fusion strategy into angular margin-based loss,which makes the trained model robust to noisy datasets.To reduce the influence of the gap between the well-labeled and noisy web data,a bridge schema is proposed via the corresponding loss that encourages the learned embeddings to be coherent.Lastly,the formulation is encapsulated into the meta-learning framework,which can reduce the overfitting of models and learn the network parameters to be noise-tolerant.Extensive experiments are performed on benchmark datasets,and the results clearly show the superiority of the proposed method over existing state-of-the-art approaches.

外文关键词：

label associated lossweighting noisy samplesfine-grained visual recognitionnoise-tolerant learningmeta-learning

作者：

Yanchao LI、Fu XIAO、Hao LI、Qun LI、Shui YU

展开 >

作者单位：

School of Computer Science,Nanjing University of Posts and Telecommunications,Nanjing 210023,China

School of Network Engineering,Zhoukou Normal University,Zhoukou 466001,China

School of Computer Science,University of Technology Sydney,Sydney 2007,Australia

基金：

National Natural Science Fund for Distinguished Young Scholars of ChinaNational Natural Science Foundation of ChinaNational Natural Science Foundation of ChinaKey Program of the National Natural Science Foundation of ChinaNatural Science Foundation of Jiangsu Province of ChinaNatural Science Foundation of Jiangsu Province of ChinaPostdoctoral Science Foundation of Jiangsu Province of ChinaChina Postdoctoral Science Foundation

项目编号：

62125203623022336227614361932013BK20180470BK202007392021K172B2021M691655

出版年：

2024

DOI：

10.1007/s11432-023-3922-2

中国科学:信息科学(英文版)

中国科学院

中国科学:信息科学(英文版)

CSTPCDEI

影响因子：0.715

ISSN：1674-733X

年,卷(期)：2024.67(6)