|
对象分割的形状共享
Shape Sharing for Object Segmentation |
|
| |
| 课程网址: |
http://videolectures.net/eccv2012_kim_shape/
|
| 主讲教师: |
Tinne Tuytelaars, Jaechul Kim, Serge J. Belongie
|
| 开课单位: |
德克萨斯大学
|
| 开课时间: |
2012-11-12
|
| 课程语种: |
英语
|
| 中文简介: |
我们为对象分割引入了类别无关的形状。现有的形状先验假设特定于类的知识，因此仅限于对象类是已知的情况。我们方法的主要见解是形状通常在不同类别的对象之间共享。为了利用这种形状共享现象，我们开发了一种非参数先验，它将对象形状从示例数据库转移到基于局部形状匹配的测试图像。然后，在图形切割公式中强制执行转移的形状先验，以产生对象段假设池。与以前的多种分割方法不同，我们的方法受益于全局形状线索;与以前的自上而下的方法不同，它假设没有特定于类的训练，并且即使对于不熟悉的类别也会增强分割。在具有挑战性的PASCAL 2010和Berkeley Segmentation数据集中，我们表明它在自下而上或类别无关分割方面优于现有技术水平。
|
| 课程简介: |
We introduce a category-independent shape prior for object segmentation. Existing shape priors assume class-specific knowledge, and thus are restricted to cases where the object class is known in advance. The main insight of our approach is that shapes are often shared between objects of different categories. To exploit this shape sharing phenomenon, we develop a non-parametric prior that transfers object shapes from an exemplar database to a test image based on local shape matching. The transferred shape priors are then enforced in a graph-cut formulation to produce a pool of object segment hypotheses. Unlike previous multiple segmentation methods, our approach benefits from global shape cues; unlike previous top-down methods, it assumes no class-specific training and thus enhances segmentation even for unfamiliar categories. On the challenging PASCAL 2010 and Berkeley Segmentation data sets, we show it outperforms the state-of-the-art in bottom-up or category independent segmentation.
|
| 关 键 词: |
对象分割; 类别无关; 形状共享
|
| 课程来源: |
视频讲座网
|
| 最后编审: |
2020-06-10：yumf
|
| 阅读次数: |
42
|
|
图 书 馆