Show simple item record

dc.contributor.advisor纪荣嵘
dc.contributor.author沈云航
dc.date.accessioned2018-12-05T01:48:11Z
dc.date.available2018-12-05T01:48:11Z
dc.date.issued2018-01-02
dc.identifier.urihttps://dspace.xmu.edu.cn/handle/2288/170630
dc.description.abstract目标检测(objectdetection)是计算机视觉领域里一个非常重要的研究问题。随着近年来深度卷积神经网络的发展,其中基于深度学习的目标检测算法在性能上取得了巨大的进步。但是目前最先进的目标检测算法需要带有精确目标物体位置标签的数据来训练模型,而这种标签信息需要花费大量人力物力来标注,同时也会引入人工标注偏差。 本文的研究内容是基于弱监督学习的目标检测问题,即没有精确的目标物体位置标签,只用图像的类别标签来学习目标检测器。基于弱监督学习的目标检测有着广泛的应用和重要的意义,也是近年来的计算机视觉领域的热门研究点。当前基于弱监督学习的目标检测算法大多是基于局部、候选区域层次的信息。对此,本...
dc.description.abstractVisual object detection is sitting on the core of computer vision research. Recently with the development of deep convolutional neural networks, object detections based on deep learning have achieved great progresses. However, current the state-of-the-art object detection algorithms require training dataset with object-level label to learn the models. It is time-consuming and labor-intensive to ob...
dc.language.isozh_CN
dc.relation.urihttps://catalog.xmu.edu.cn/opac/openlink.php?strText=58800&doctype=ALL&strSearchType=callno
dc.source.urihttps://etd.xmu.edu.cn/detail.asp?serial=60233
dc.subject弱监督学习
dc.subject目标检测
dc.subject深度学习
dc.subjectWeakly Supervised Learning
dc.subjectObject Detection
dc.subjectDeep Learning
dc.title基于特定类别空间约束的弱监督目标检测
dc.title.alternativeCategory-Aware Spatial Constraint for Weakly Supervised Object Detection
dc.typethesis
dc.date.replied2017-05-18
dc.description.note学位:工程硕士
dc.description.note院系专业:信息科学与技术学院_工程硕士(计算机技术)
dc.description.note学号:31520141153305


Files in this item

Thumbnail

This item appears in the following Collection(s)

Show simple item record