arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 2009.07536cs.CV
收藏

基于多分辨特征的混合注意力引导网络的身份再识别

Hybrid-Attention Guided Network with Multiple Resolution Features for Person Re-Identification

Nanjing University of Information Science and Technology(南京信息工程大学);Nanjing University of Finance and Economics(南京财经大学);Wormpex AI Research(Wormpex AI研究院);Tianjin University of Technology(天津理工大学)

Guoqing Zhang, Junchuan Yang, Yuhui Zheng, Yi Wu, Shengyong Chen

摘要:

提取有效的和有区别的特征对于解决具有挑战性的人重新识别(re-ID)任务非常重要。流行的深度卷积神经网络(CNN)通常使用高级特征来识别行人。然而,由于在训练阶段大量的填充和池化操作,在学习高层特征时,一些基本的空间信息会丢失在低层特征中,如形状、纹理和颜色。此外,大多数现有的人重新ID方法主要是基于手工制作的边界框,其中图像被精确对齐。在实际应用中,由于现有的目标检测算法往往产生不准确的边界框,这是不现实的。这将不可避免地降低现有算法的性能。为了解决这些问题,我们提出了一种新的人重新识别模型,融合高,低层次的嵌入,以减少在学习高层次的功能造成的信息丢失。然后将融合后的嵌入分割为若干部分,并将其重新连接,以获得全局特征和更重要的局部特征,从而减轻不准确的包围盒对嵌入效果的影响。此外,我们还在我们的模型中引入了空间和通道的注意力机制,旨在挖掘与目标相关的更具区分性的特征。最后,我们重新构造了特征提取器,以确保我们的模型可以获得更丰富和鲁棒的特征。大量的实验表明,我们的方法相比,现有的方法的优越性。我们的代码可在https://github.com/libraflower/MutipleFeature-for-PRID上获得。

英文摘要:

Extracting effective and discriminative features is very important for addressing the challenging person re-identification (re-ID) task. Prevailing deep convolutional neural networks (CNNs) usually use high-level features for identifying pedestrian. However, some essential spatial information resided in low-level features such as shape, texture and color will be lost when learning the high-level features, due to extensive padding and pooling operations in the training stage. In addition, most existing person re-ID methods are mainly based on hand-craft bounding boxes where images are precisely aligned. It is unrealistic in practical applications, since the exploited object detection algorithms often produce inaccurate bounding boxes. This will inevitably degrade the performance of existing algorithms. To address these problems, we put forward a novel person re-ID model that fuses high- and low-level embeddings to reduce the information loss caused in learning high-level features. Then we divide the fused embedding into several parts and reconnect them to obtain the global feature and more significant local features, so as to alleviate the affect caused by the inaccurate bounding boxes. In addition, we also introduce the spatial and channel attention mechanisms in our model, which aims to mine more discriminative features related to the target. Finally, we reconstruct the feature extractor to ensure that our model can obtain more richer and robust features. Extensive experiments display the superiority of our approach compared with existing approaches. Our code is available at https://github.com/libraflower/MutipleFeature-for-PRID.

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。

↑