首页 文献索引 SCI期刊 AI助手
登录 注册
首页 正文

Advanced intelligent systems (Weinheim an der Bergstrasse, Germany). 2024 Sep;6(9):2400044. doi: 10.1002/aisy.202400044 Q16.82024

Deformable Capsules for Object Detection

用于目标检测的可变形胶囊 翻译改进

Rodney LaLonde  1, Naji Khosravan  2, Ulas Bagci  3

作者单位 +展开

作者单位

  • 1 Palantir Technologies, Washington, DC.
  • 2 Zillow, Seattle, WA.
  • 3 Northwestern University, Chicago, IL.
  • DOI: 10.1002/aisy.202400044 PMID: 39669747

    摘要 中英对照阅读

    Capsule networks promise significant benefits over convolutional networks by storing stronger internal representations, and routing information based on the agreement between intermediate representations' projections. Despite this, their success has been limited to small-scale classification datasets due to their computationally expensive nature. Though memory efficient, convolutional capsules impose geometric constraints that fundamentally limit the ability of capsules to model the pose/deformation of objects. Further, they do not address the bigger memory concern of class-capsules scaling up to bigger tasks such as detection or large-scale classification. In this study, we introduce a new family of capsule networks, deformable capsules (DeformCaps), to address a very important problem in computer vision: object detection. We propose two new algorithms associated with our DeformCaps: a novel capsule structure (SplitCaps), and a novel dynamic routing algorithm (SE-Routing), which balance computational efficiency with the need for modeling a large number of objects and classes, which have never been achieved with capsule networks before. We demonstrate that the proposed methods efficiently scale up to create the first-ever capsule network for object detection in the literature. Our proposed architecture is a one-stage detection framework and it obtains results on MS COCO which are on par with state-of-the-art one-stage CNN-based methods, while producing fewer false positive detection, generalizing to unusual poses/viewpoints of objects.

    Keywords: Capsule networks; SE-Routing; SplitCaps; deformable capsules; large-scale classification; object detection.

    Keywords:deformable capsules; object detection

    胶囊网络通过存储更强的内部表示和基于中间表示投影之间一致性的路由信息,有望比卷积网络具有显著的优势。尽管如此,由于其计算成本高昂的性质,它们的成功仅限于小规模的分类数据集。尽管卷积胶囊具有内存效率,但它施加了几何约束,从根本上限制了胶囊对物体姿态/变形进行建模的能力。此外,它们没有解决类胶囊扩展到更大任务(如检测或大规模分类)的更大内存问题。在这项研究中,我们引入了一个新的胶囊网络家族,可变形胶囊(DeformaCaps),以解决计算机视觉中一个非常重要的问题:物体检测。我们提出了两种与DeformCaps相关的新算法:一种新型的胶囊结构(SplitCaps)和一种新型动态路由算法(SE routing),它们在计算效率与对大量对象和类进行建模的需求之间取得了平衡,这是胶囊网络以前从未实现过的。我们证明,所提出的方法可以有效地扩展到创建文献中第一个用于目标检测的胶囊网络。我们提出的架构是一个单阶段检测框架,它在MS COCO上获得的结果与最先进的基于CNN的单阶段方法相当,同时产生的误报检测更少,适用于物体的异常姿势/视点。关键词:胶囊网络;SE路由;拼接封套;可变形胶囊;大规模分类;物体检测。

    关键词:可变形胶囊; 目标检测

    翻译效果不满意? 用Ai改进或 寻求AI助手帮助 ,对摘要进行重点提炼
    Copyright © Advanced intelligent systems (Weinheim an der Bergstrasse, Germany). 中文内容为AI机器翻译,仅供参考!

    相关内容

    期刊名:Advanced intelligent systems

    缩写:

    ISSN:2640-4567

    e-ISSN:

    IF/分区:6.8/Q1

    文章目录 更多期刊信息

    全文链接
    引文链接
    复制
    已复制!
    推荐内容
    Deformable Capsules for Object Detection