Fine-Grained 3D-Attention Prototypes for Few-Shot Learning

Fine-Grained 3D-Attention Prototypes for Few-Shot Learning
复制标题

用于 FewShot 学习的细粒度 3D 注意力原型

DOI:
10.1162/neco_a_01302
复制
发表时间:
2020-07
期刊:
影响因子:
2.9
通讯作者:
Yudai Pan
Yudai Pan
中科院分区:
计算机科学4区
文献类型:
--
作者:
Xin Hu;Jun Liu;Jie Ma;Yudai Pan

文献摘要

参考文献

相似文献

在真实的世界中,每个类别有限数量的标记细粒度图像很难有效地表示类别分布。由于细粒度图像比具有明显对象的简单图像具有更细微的视觉差异,即存在更小的类间差异和更大的类内差异。为了解决这些问题,我们提出了一个端到端的基于注意力的细粒度少镜头图像分类(AFG)模型与最近的情节训练策略。它主要由特征学习模块、图像重建模块和标签分配模块组成。特征学习模块主要设计了一种3D-Attention机制,该机制同时考虑了图像特征的空间位置和不同通道的关注度,以学习更多有区别的局部特征,更好地表示类别分布。图像重建模块计算局部特征与原始图像之间的映射。该方法通过设计一个损失函数作为辅助监督信息,使得每个局部特征的学习不需要额外的标注。标签分布模块用于预测给定未标记样本的标签分布,并使用局部特征来表示图像特征以进行分类。通过在Mini-ImageNet和三个细粒度数据集上进行综合实验,我们证明了所提出的模型比竞争对手具有上级性能。
In the real world, a limited number of labeled finely grained images per class can hardly represent the class distribution effectively. Due to the more subtle visual differences in fine-grained images than simple images with obvious objects, that is, there exist smaller interclass and larger intraclass variations. To solve these issues, we propose an end-to-end attention-based model for fine-grained few-shot image classification (AFG) with the recent episode training strategy. It is composed mainly of a feature learning module, an image reconstruction module, and a label distribution module. The feature learning module mainly devises a 3D-Attention mechanism, which considers both the spatial positions and different channel attentions of the image features, in order to learn more discriminative local features to better represent the class distribution. The image reconstruction module calculates the mappings between local features and the original images. It is constrained by a designed loss function as auxiliary supervised information, so that the learning of each local feature does not need extra annotations. The label distribution module is used to predict the label distribution of a given unlabeled sample, and we use the local features to represent the image features for classification. By conducting comprehensive experiments on Mini-ImageNet and three fine-grained data sets, we demonstrate that the proposed model achieves superior performance over the competitors.
DOI: 10.1109/cvpr.2019.00515
发表时间: 2019-03
期刊: 2019 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR)
影响因子: --
作者:
Heliang Zheng;Jianlong Fu;Zhengjun Zha;Jiebo Luo
通讯作者: Heliang Zheng;Jianlong Fu;Zhengjun Zha;Jiebo Luo
DOI: 10.1109/tmm.2015.2408566
发表时间: 2015-03
影响因子: 7.3
作者:
Lingxi Xie;Jingdong Wang;Bo Zhang-;Qi Tian
通讯作者: Lingxi Xie;Jingdong Wang;Bo Zhang-;Qi Tian
通过匹配网络进行一次性学习的计划采样
DOI: 10.1016/j.patcog.2019.07.007
发表时间: 2019-12
影响因子: 8
作者:
Lingling Zhang;Jun Liu;Minnan Luo
通讯作者: Minnan Luo
DOI: --
发表时间: 2011-07
期刊: --
影响因子: --
作者:
C. Wah;Steve Branson;P. Welinder;P. Perona;Serge J. Belongie
通讯作者: C. Wah;Steve Branson;P. Welinder;P. Perona;Serge J. Belongie
DOI: --
发表时间: 2017
期刊: --
影响因子: --
作者:
通讯作者: --