arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 1604.03239cs.CV
收藏

图像中的CRAFT对象

CRAFT Objects from Images

Institute of Automation, Chinese Academy of Sciences(中国科学院自动化研究所);Tsinghua University(清华大学)

Bin Yang, Junjie Yan, Zhen Lei, Stan Z. Li

摘要:

目标检测是图像理解中的一个基本问题。一个流行的解决方案是R-CNN框架及其快速版本。他们将目标检测问题分解为两个级联的更简单的任务:1)从图像中生成目标建议,2)将建议分类为各种目标类别。尽管我们正在处理两个相对容易的任务,但它们并没有完美地解决,仍然有改进的空间。在本文中,我们通过将每个任务划分为两个子任务,进一步推动了“分而治之”的解决方案。我们称所提出的方法为“CRAFT”(级联区域建议网络和FasT-rcnn),它处理每个任务与精心设计的网络级联。我们表明,级联结构有助于这两项任务:在建议生成,它提供了更紧凑,更好地本地化的对象建议;在对象分类,它通过捕获类别间和类别内的差异,减少误报(主要是在模糊的类别之间)。CRAFT在目标检测基准测试(如PASCAL VOC 07/12和ILSVRC)上取得了一致且显著的改进。

英文摘要:

Object detection is a fundamental problem in image understanding. One popular solution is the R-CNN framework and its fast versions. They decompose the object detection problem into two cascaded easier tasks: 1) generating object proposals from images, 2) classifying proposals into various object categories. Despite that we are handling with two relatively easier tasks, they are not solved perfectly and there's still room for improvement. In this paper, we push the "divide and conquer" solution even further by dividing each task into two sub-tasks. We call the proposed method "CRAFT" (Cascade Region-proposal-network And FasT-rcnn), which tackles each task with a carefully designed network cascade. We show that the cascade structure helps in both tasks: in proposal generation, it provides more compact and better localized object proposals; in object classification, it reduces false positives (mainly between ambiguous categories) by capturing both inter- and intra-category variances. CRAFT achieves consistent and considerable improvement over the state-of-the-art on object detection benchmarks like PASCAL VOC 07/12 and ILSVRC.

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。

↑