1606.End--to-End Comparative Attention Networks for Person Re-identification 论文笔记
2017-12-31 11:13
661 查看
简介
参考人类视觉机制:即重复性的抓住某一区域主要信息来发现比较对象的相似性和差异性,来发现最匹配的目标,
利用LSTM网络设计注意力机制,增加了时空信息,设计端到端训练的比较性注意力网络(CAN),训练阶段,该网络采用online triplet损失,可以使LSTM网络部分对同一ID的对个显著性局部区域比较,并pull他们之间的相似性,而对不同ID的两个行人的多个显著性区域也进行学习,使之push开,得到大的表观特征区分性。
end to end 的说法:
This framework simulates the re-identification process of human visual system by learning a comparative model from raw person images to recurrently localize some discriminative parts of person images via a set of glimpses. At each glimpse, the model generates different parts without any manual annotations. And,this model generates attention regions based on the pre-extracted CNN.
features.
提取的特征类似于一种动态特征pooling的操作
The end-to-end CAN learns to selectively focus on parts of pairs of person images after taking a few glimpses of them and adaptively comparing their appearance.
在测试阶段,则所proposed CAN model simulates the human perception process to verify whether two images are from the same person.
模型架构:
global discriminative feature learning components,采用截断的CNN 网络,如Alexnet或VGG,训练时,先用softmax预训练一个classification model,然后移除后面的三个全连接层,他们被替换为LSTM网络
comparative attention components,一般用LSTM网络,产生the comparative visual attention regional features
作者设计了一个concatenation layer is applied to concatenate a few time steps of hidden states along channel axis来整合某几个时间步的特征,
之后,采用多任务的训练模式:triplet loss + identification loss,jointly end-to-end train our architecture with them,然后对特征进行正则化:
多loss结合的形式:
test 训练好的网络:图相对输入+欧氏距离计算+ranking unit
关键词:Comparative Attention Network, Multiple glimpses
作者的模型可以自适应的发现多个具有更多区分性信息的局部区域,这可以进一步提升reid的性能。
改进方向:考虑尺度变化问题。
相关文章推荐
- 【论文笔记】Person Transfer GAN to Bridge Domain Gap for Person Re-Identification
- 论文笔记《End-to-End Deep Learning for Person Search》
- 1703.In Defense of the Triplet Loss for Person Re-Identification 论文阅读笔记
- 【论文笔记】An End-to-End Model for QA over KBs with Cross-Attention Combining Global Knowledge
- 【论文笔记】Margin Sample Mining Loss: A Deep Learning Based Method for Person Re-identification
- part-aligned系列论文:1707.Deeply-Learned Part-Aligned Representations for Person Re-Identification 论文笔记
- 1505.Deep Ranking for Person Re-identification via Joint Representation Learning论文笔记
- 【论文笔记】In Defense of the Triplet Loss for Person Re-Identification
- 论文笔记:Improving Deep Visual Representation for Person Re-identification by Global and Local Image-language Association
- 论文阅读:Multi-Scale Triplet CNN for Person Re-Identification
- Query-Adaptive Late Fusion for Image Search and Person Re-identification阅读笔记
- 1504.ICCVPartial Person Re-Identification 论文笔记
- 论文阅读:Joint Learning of Single-image and Cross-image Representations for Person Re-identification
- 论文阅读:End-to-End Learning of Deformable Mixture of Parts and Deep Convolutional Neural Networks for H
- 论文笔记(一)Deep Ranking for Person Re-Identification via Joint Representation Learning (续)
- [论文笔记] An End-to-End Approach for QoS-Aware Service Composition (EDOC, 2009)
- 论文笔记之:Let there be Color!: Joint End-to-end Learning of Global and Local Image Priors for Automatic
- 论文笔记之:Fully Convolutional Attention Localization Networks: Efficient Attention Localization for Fine-Grained Recognition
- 论文笔记之---Joint Detection and Identification Feature Learning for Person Search