当前位置: X-MOL 学术Int. J. Pattern Recognit. Artif. Intell. › 论文详情
Our official English website, www.x-mol.net, welcomes your feedback! (Note: you will need to create a separate account there.)
Baggage Image Retrieval with Attention-Based Network for Security Checks
International Journal of Pattern Recognition and Artificial Intelligence ( IF 0.9 ) Pub Date : 2021-05-14 , DOI: 10.1142/s0218001421550090
Gan Huang 1 , Li Yang 1 , Ding Zhang 2 , Xiaofeng Wang 1 , Yanfu Wang 1
Affiliation  

Baggage image search is the task of finding the same baggage across several cameras, which can improve the efficiency of security check. Since existing state-of-the-art image retrieval requires a significant amount of training data, we aim to investigate the few-shot situation, utilizing only a few samples for each baggage. In this paper, the framework we introduced is called AttentionNet, exploring the weak region-wise annotations as attention clues to improve the retrieval performance. Specifically, we integrate the semantic and recognition tasks using shared convolutional neural networks by multi-task learning. Semantic attention mechanism is used to explicitly re-weight the feature map for emphasizing the foreground. To prevent overfitting in few-shot training, we adopt a variant of the triplet loss to perform deep metric learning with an online hard triplet mining strategy. Once trained, AttentionNet identifies a probe image by computing its embedding’s cosine distance with images in the gallery. Experiments show that our method achieves state-of-the-art accuracy in image search. For instance, we obtain 82.9% Rank-1 score on the MVB dataset, dramatically outperforming the currently existing methods.

中文翻译:

基于注意力网络的行李图像检索以进行安全检查

行李图像搜索是跨多个摄像头查找相同行李的任务,可以提高安检效率。由于现有的最先进的图像检索需要大量的训练数据,我们的目标是调查少样本的情况,每个行李只使用几个样本。在本文中,我们介绍的框架称为 AttentionNet,探索弱区域标注作为注意力线索,以提高检索性能。具体来说,我们通过多任务学习使用共享卷积神经网络集成语义和识别任务。语义注意机制用于显式地重新加权特征图以强调前景。为了防止在少样本训练中过度拟合,我们采用三元组损失的一种变体,通过在线硬三元组挖掘策略执行深度度量学习。训练完成后,AttentionNet 通过计算其嵌入与图库中图像的余弦距离来识别探测图像。实验表明,我们的方法在图像搜索中达到了最先进的精度。例如,我们在 MVB 数据集上获得了 82.9% 的 Rank-1 分数,大大优于目前现有的方法。
更新日期:2021-05-14
down
wechat
bug