arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 2009.01438cs.CVcs.AI
收藏

任务集成网络:图像搜索的联合检测与检索

Tasks Integrated Networks: Joint Detection and Retrieval for Image Search

Chongqing University(重庆大学);University of Technology Sydney(悉尼科技大学);Institute of Automation, Chinese Academy of Sciences(中国科学院自动化研究所);Chongqing University of Posts and Telecommunications(重庆邮电大学)

Lei Zhang, Zhenwei He, Yi Yang, Liang Wang, Xinbo Gao

摘要:

传统的对象检索任务旨在学习具有内部相似性和内部不相似性的区别性特征表示,这假设图像中的对象被手动或自动精确地预裁剪。然而,在许多现实世界的搜索场景中(例如,视频监视),对象(例如,人员、车辆等)很少被准确地检测或注释。因此,对象级检索变得棘手,没有边界框注释,这导致了一个新的,但具有挑战性的课题,即图像级搜索。本文针对图像搜索问题,首先介绍了一种端到端的集成网络(I-Net),它具有以下三个优点:1)设计了一种Siamese体系结构和图像中相似和不相似对象的在线配对策略。2)引入了一种新的在线配对(OLP)损失与一个动态的特征字典,它通过自动生成一些负面的对,以限制积极的多任务训练停滞问题。3)提出了一种基于软最大损失的硬样本优先级(HEP)算法,通过选择硬类别来提高分类任务的鲁棒性。基于分而治之的思想,我们进一步提出了一种改进的I-Net,称为DC-I-Net,它有两个新的贡献:1)在集成框架中裁剪了两个模块来分别处理不同的任务,从而保证了任务的规范性。2)提出了一种基于类中心的HEP丢失算法(C2 HEP),该算法利用存储的类中心来捕获类内相似性和类间不相似性,从而实现最终的检索。在著名的面向图像级搜索的基准数据集上的大量实验表明,所提出的DC-I-Net优于最先进的任务集成和任务分离的图像搜索模型。

英文摘要:

The traditional object retrieval task aims to learn a discriminative feature representation with intra-similarity and inter-dissimilarity, which supposes that the objects in an image are manually or automatically pre-cropped exactly. However, in many real-world searching scenarios (e.g., video surveillance), the objects (e.g., persons, vehicles, etc.) are seldom accurately detected or annotated. Therefore, object-level retrieval becomes intractable without bounding-box annotation, which leads to a new but challenging topic, i.e. image-level search. In this paper, to address the image search issue, we first introduce an end-to-end Integrated Net (I-Net), which has three merits: 1) A Siamese architecture and an on-line pairing strategy for similar and dissimilar objects in the given images are designed. 2) A novel on-line pairing (OLP) loss is introduced with a dynamic feature dictionary, which alleviates the multi-task training stagnation problem, by automatically generating a number of negative pairs to restrict the positives. 3) A hard example priority (HEP) based softmax loss is proposed to improve the robustness of classification task by selecting hard categories. With the philosophy of divide and conquer, we further propose an improved I-Net, called DC-I-Net, which makes two new contributions: 1) two modules are tailored to handle different tasks separately in the integrated framework, such that the task specification is guaranteed. 2) A class-center guided HEP loss (C2HEP) by exploiting the stored class centers is proposed, such that the intra-similarity and inter-dissimilarity can be captured for ultimate retrieval. Extensive experiments on famous image-level search oriented benchmark datasets demonstrate that the proposed DC-I-Net outperforms the state-of-the-art tasks-integrated and tasks-separated image search models.

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。

↑