基于增强特征融合网络的行人再识别

针对行人再识别问题,目前多数方法将行人的局部或全局特征分开考虑,从而忽略了行人整体之间的关系,即行人全局特征和局部特征之间的联系。本文提出一种增强特征融合网络(enhanced feature convergent network,EFCN)。在全局分支中,提出适用于获取全局特征的注意力网络作为嵌入特征,嵌入在基础网络模型中以提取行人的全局特征;在局部分支中,提出循环门单元变换网络(gated recurrent unit change network,GRU-CN)得到代表性的局部特征;再使用特征融合方法将全局特征和局部特征融合成最终的行人特征;最后借助损失函数训练网络。通过大量的对比实验表明,该算法网络模型在标准的Re-ID数据集上可以获得较好的实验结果。提出的增强特征融合网络能提取辨别性较强的行人特征,该模型能够应用于大场景非重叠多摄像机下的行人再识别问题,具有较高的识别能力和识别精度,且对背景变化的行人图像能提取具有较强的鲁棒性特征。

Enhanced feature convergent network for person re-identification

For the problem of person re-identification most of the current methods only considered the local or global features of the pedestrian separately,ignoring the relationship between the features.This paper proposed the enhanced feature convergent network(EFCN).In the global branch,the paper proposed the attention network suitable for obtained global features as the embedded feature,which was embedded in the basic network model to extract global feature of pedestrians.In the local branch,it proposed the gated recurrent unit change network(GRU-CN)to obtain more robust local features.Then this paper used feature fusion to connect the extracted global and local features.Extensive comparative experiments show that EFCN can achieve better experimental results on three standard person Re-ID datasets.The EFCN can extract highly discriminative pedestrian features.This model can be applied to the problem of Re-ID under non-overlapping multi-cameras in large scenes,which has high recognition ability and accuracy.The method can extract robust features for pedestrian images with changing background.