Knowledge Network Node
Journal of Chongqing Technology and Business University(Natural Science Edition).
(Accepted Version) Online First Publishing Date: 2025-03-10 08:41:10

Multi-Label Classification Based on Semantic Space Attention and Contrast LearningChinese Full Text

Wang Sunbin;Wang Dian;Huang Jun;

Abstract: Objective: Multi-label image classification aims to assign corresponding labels to multiple objects within an image. One of the key challenges in multi-label image classification lies in explore and utilize the potential correlations between labels. Existing methods based on graph neural networks for modeling label relationships in multi-label classification typically neglect the intra-class and inter-class relationships between specific semantic features. Additionally， label noise and label co-... More
Keywords：

Multi-label image classification; Label correlation; Label co-occurrence; Attention mechanism; Contrastive learning;

  • Series：

    (I) Electronic Technology ＆ Information Science

  • Subject：

    Computer Software and Application of Computer; Automation Technology

  • Classification Code：

    TP391.41;TP18

CNKI exclusive online-first articles is prohibited to reprint and excerpt without permission.

Download the mobile appuse the app to scan this coderead the article.

Tips: Please download CAJViewer to view CAJ format full text.

Download: 100 Pagecount: 14 Size: 1342K

Related Literature
  • Similar Article
  • Reader Recommendation
  • Associated Author