![一种用于显著性前景检测和多对象分割的迭代法](/CN/2017/8/15/images/201780078605.jpg)
基本信息:
- 专利标题: 一种用于显著性前景检测和多对象分割的迭代法
- 专利标题(英):ITERATIVE METHOD FOR SALIENT FOREGROUND DETECTION AND MULTI-OBJECT SEGMENTATION
- 申请号:CN201780078605.4 申请日:2017-12-19
- 公开(公告)号:CN110088805A 公开(公告)日:2019-08-02
- 发明人: 亚历山大·路易 , 大卫·克洛斯特曼 , 麦加·库塞 , 内森·卡希尔 , 大卫·梅辛格
- 申请人: 柯达阿拉里斯股份有限公司
- 申请人地址: 美国纽约
- 专利权人: 柯达阿拉里斯股份有限公司
- 当前专利权人: 柯达阿拉里斯股份有限公司
- 当前专利权人地址: 美国纽约
- 代理机构: 成都超凡明远知识产权代理有限公司
- 代理人: 魏彦; 洪玉姬
- 优先权: 62/436,803 2016.12.20 US
- 国际申请: PCT/US2017/067309 2017.12.19
- 国际公布: WO2018/118914 EN 2018.06.28
- 进入国家日期: 2019-06-19
- 主分类号: G06T7/11
- IPC分类号: G06T7/11
A system and a method that perform iterative foreground detection and multi-object segmentation in an image are disclosed herein. A new background prior is introduced to improve the foreground segmentation results. Three complimentary methods detect and segment foregrounds containing multiple objects. The first method performs an iterative segmentation of the image to pull out the salient objectsin the image. In a second method, a higher dimensional embedding of the image graph is used to estimate the saliency score and extract multiple salient objects. A third method uses a metric to automatically pick the number of eigenvectors to consider in an alternative method to iteratively compute the image saliency map. Experimental results show that these methods succeed in accurately extractingmultiple foreground objects from an image.
公开/授权文献:
- CN110088805B 用于检测和分割多个前景对象的方法 公开/授权日:2023-06-06