Deep Large-Margin Rank Loss for Multi-Label Image Classification

Author:

Ma Zhongchen,Li Zongpeng,Zhan Yongzhao

Abstract

The large-margin technique has served as the foundation of several successful theoretical and empirical results in multi-label image classification. However, most large-margin techniques are only suitable to shallow multi-label models with preset feature representations and a few large-margin techniques of neural networks only enforce margins at the output layer, which are not well suitable for deep networks. Based on the large-margin technique, a deep large-margin rank loss function suitable for any network structure is proposed, which is able to impose a margin on any chosen set of layers of a deep network, allows choosing any ℓp norm (p≥1) on the metric measuring the margin between labels and is applicable to any network architecture. Although the complete computation of deep large-margin rank loss function has the O(C2) time complexity, where C denotes the size of the label set, which would cause scalability issues when C is large, a negative sampling technique was proposed to make the loss function scale linearly to C. Experimental results on two large-scale datasets, VOC2007 and MS-COCO, show that the deep large-margin ranking function improves the robustness of the model in multi-label image classification tasks while enhancing the model’s anti-noise performance.

Funder

National Natural Science Foundation of China

China Postdoctoral Science Foundation

Publisher

MDPI AG

Subject

General Mathematics,Engineering (miscellaneous),Computer Science (miscellaneous)

Reference41 articles.

1. Deep integration: A multi-label architecture for road scene recognition;Chen;IEEE Trans. Image Process.,2019

2. Semantic-interactive graph convolutional network for multilabel image recognition;Chen;IEEE Trans. Syst. Man Cybern. Syst.,2021

3. Ge, Z., Mahapatra, D., Sedai, S., Garnavi, R., and Chakravorty, R. (2018). Chest x-rays classification: A multi-label and fine-grained problem. arXiv.

4. Multilabel classification of medical concepts for patient clinical profile identification;Vaillant;Artif. Intell. Med.,2022

5. Learning multi-label scene classification;Boutell;Pattern Recognit.,2004

Cited by 2 articles. 订阅此论文施引文献 订阅此论文施引文献,注册后可以免费订阅5篇论文的施引文献,订阅后可以查看论文全部施引文献

同舟云学术

1.学者识别学者识别

2.学术分析学术分析

3.人才评估人才评估

"同舟云学术"是以全球学者为主线,采集、加工和组织学术论文而形成的新型学术文献查询和分析系统,可以对全球学者进行文献检索和人才价值评估。用户可以通过关注某些学科领域的顶尖人物而持续追踪该领域的学科进展和研究前沿。经过近期的数据扩容,当前同舟云学术共收录了国内外主流学术期刊6万余种,收集的期刊论文及会议论文总量共计约1.5亿篇,并以每天添加12000余篇中外论文的速度递增。我们也可以为用户提供个性化、定制化的学者数据。欢迎来电咨询!咨询电话:010-8811{复制后删除}0370

www.globalauthorid.com

TOP

Copyright © 2019-2024 北京同舟云网络信息技术有限公司
京公网安备11010802033243号  京ICP备18003416号-3