###
计算机系统应用英文版:2024,33(9):48-57
本文二维码信息
码上扫一扫!
基于特征解耦和开放性学习的小样本开放集识别
(江西理工大学 信息工程学院, 赣州 341000)
Few-shot Open-set Recognition with Feature Decoupling and Openness Learning
(School of Information Engineering, Jiangxi University of Science and Technology, Ganzhou 341000, China)
摘要
图/表
参考文献
相似文献
本文已被:浏览 219次   下载 1656
Received:March 07, 2024    Revised:April 10, 2024
中文摘要: 在小样本开放集识别任务中, 有效区分闭集和开集样本是一项具有挑战性的任务, 尤其在样本稀缺的情况下. 现有方法在构造已知类分布边界时存在模糊性, 未能很好地实现闭集和开集空间的区分. 为了解决这一问题, 本文提出了基于特征解耦和开放性学习的小样本开放集识别方法. 其目的是通过特征解耦模块, 迫使模型解耦类别性特征和开放性特征, 从而扩大未知类与已知类之间的差异. 为了有效实现特征解耦, 引入了开放性学习损失来促进特征的开放性学习. 通过结合相似度度量值和反开放性分值作为损失优化对象, 更好地引导模型学习到更具区分性的特征表示. 实验结果表明, 本文方法在公共数据集miniImageNet和tieredImageNet上可以显著提高未知类样本的检测率, 同时正确分类已知类别.
Abstract:In the task of few-shot open-set recognition (FSOSR), effectively distinguishing closed-set from open-set samples presents a notable challenge, especially in cases of sample scarcity. Current approaches exhibit uncertainty in describing boundaries for known class distributions, leading to insufficient discrimination between closed-set and open-set spaces. To tackle this issue, this study introduces a novel method for FSOSR leveraging feature decoupling and openness learning. The primary objective is to employ a feature decoupling module to compel the model to decouple class-specific features and open-set features, thereby accentuating the disparity between unknown and known classes. To achieve effective feature decoupling, an openness learning loss is introduced to facilitate the acquisition of open-set features. By integrating similarity metric values and anti-openness scores as the optimization target, the model is steered towards learning more discriminative feature representations. Experimental results on publicly datasets miniImageNet and tieredImageNet demonstrate that the proposed method substantially enhances the detection rate of unknown class samples while accurately classifying known classes.
文章编号:     中图分类号:    文献标志码:
基金项目:国家自然科学基金(62361032); 江西省主要学科学术和技术带头人领军人才项目(20213BCJ22004); 江西省自然科学基金重点项目(20232ACB202011)
引用文本:
吴少玲,罗会兰.基于特征解耦和开放性学习的小样本开放集识别.计算机系统应用,2024,33(9):48-57
WU Shao-Ling,LUO Hui-Lan.Few-shot Open-set Recognition with Feature Decoupling and Openness Learning.COMPUTER SYSTEMS APPLICATIONS,2024,33(9):48-57