arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 2309.07186cs.CV
收藏

LCReg:基于潜在类别的识别的长尾图像分类

LCReg: Long-Tailed Image Classification with Latent Categories based Recognition

Institute for Infocomm Research, A*STAR(新加坡科技研究局资讯通信研究院);Nanyang Technological University(南洋理工大学)

Weide Liu, Zhonghua Wu, Yiming Wang, Henghui Ding, Fayao Liu, Jie Lin, Guosheng Lin

AI总结:

针对长尾图像识别中尾部类别样本少导致特征表示不佳的问题,提出LCReg方法,通过学习头部和尾部类别共享的类别无关潜在特征并进行语义数据增强,显著提升了基线性能。

摘要:

在这项工作中,我们解决长尾图像识别这一具有挑战性的问题。以往的长尾识别方法主要关注对尾部类别的数据增强或重平衡策略,以在模型训练时给予它们更多关注。然而,这些方法受限于尾部类别的训练图像数量少,导致特征表示不佳。为解决此问题,我们提出了基于潜在类别的长尾识别方法(LCReg)。我们的假设是,头部和尾部类别共享的常见潜在特征可用于改善特征表示。具体而言,我们学习一组头部和尾部类别共享的类别无关潜在特征,然后对潜在特征进行语义数据增强,以隐式增加训练样本的多样性。我们在五个长尾图像识别数据集上进行了大量实验,结果表明我们提出的方法显著改善了基线。

英文摘要:

In this work, we tackle the challenging problem of long-tailed image recognition. Previous long-tailed recognition approaches mainly focus on data augmentation or re-balancing strategies for the tail classes to give them more attention during model training. However, these methods are limited by the small number of training images for the tail classes, which results in poor feature representations. To address this issue, we propose the Latent Categories based long-tail Recognition (LCReg) method. Our hypothesis is that common latent features shared by head and tail classes can be used to improve feature representation. Specifically, we learn a set of class-agnostic latent features shared by both head and tail classes, and then use semantic data augmentation on the latent features to implicitly increase the diversity of the training sample. We conduct extensive experiments on five long-tailed image recognition datasets, and the results show that our proposed method significantly improves the baselines.

机构、总结、中文摘要均由AI分析而来,可能存在错误,请以论文正文为准。

↑