Mean Shift A Robust Approach to Feature Space Aalysis均值漂移到特征空间分析的一个有力的方法.ppt

下载文档 降价啦

8
0
约5.59千字
约 46页
2018-06-19 发布于福建
举报
版权申诉
保障服务

Mean Shift A Robust Approach to Feature Space Aalysis均值漂移到特征空间分析的一个有力的方法.ppt

1、本文档共46页，可阅读全部内容。
2、原创力文档（book118）网站文档一经付费（服务费），不意味着购买了该文档的版权，仅供个人/单位学习、研究之用，不得用于商业用途，未经授权，严禁复制、发行、汇编、翻译或者网络传播等，侵权必究。
3、本站所有内容均由合作方或网友上传，本站不对文档的完整性、权威性及其观点立场正确性做任何保证或承诺！文档内容仅供研究参考，付费前请自行鉴别。如您付费，意味着您自己接受本站规则且自行承担风险，本站不退款、不进行额外附加服务；查看《如何避免下载的几个坑》。如果您已付费下载过本站文档，您可以点击这里二次下载。
4、如文档侵犯商业秘密、侵犯著作权、侵犯人身权等，请点击“版权申诉”（推荐），也可以打举报电话：400-050-0827(电话支持时间：9:00-18:30)。

Mean Shift A Robust Approach to Feature Space Aalysis均值漂移到特征空间分析的一个有力的方法

Mean ShiftA Robust Approach toFeature Space Analysis Kalyan Sunkavalli 04/29/2008 ES251R An Example Feature Space An Example Feature Space An Example Feature Space Mean Shift A non-parametric technique for analyzing complex multimodal feature spaces and estimating the stationary points (modes) of the underlying probability density function without explicitly estimating it. Outline Mean Shift An intuition Kernel Density Estimation Derivation Properties Applications of Mean Shift Discontinuity preserving Smoothing Image Segmentation Outline Mean Shift An intuition Kernel Density Estimation Derivation Properties Applications of Mean Shift Discontinuity preserving Smoothing Image Segmentation Intuitive Description Intuitive Description Intuitive Description Intuitive Description Intuitive Description Intuitive Description Intuitive Description Outline Mean Shift An intuition Kernel Density Estimation Derivation Properties Applications of Mean Shift Discontinuity preserving Smoothing Image Segmentation Parzen Windows Kernel Properties Bounded Compact support Normalized Symmetric Exponential decay Kernels and Bandwidths Kernel Types Bandwidth Parameter Various Kernels Outline Mean Shift An intuition Kernel Density Estimation Derivation Properties Applications of Mean Shift Discontinuity preserving Smoothing Image Segmentation Density Gradient Estimation Mean Shift Mean Shift Mean Shift is proportional to the normalized density gradient estimate obtained with kernel The normalization is by the density estimate computed with kernel Outline Mean Shift An intuition Kernel Density Estimation Derivation Properties Applications of Mean Shift Discontinuity preserving Smoothing Image Segmentation Properties of Mean Shift Guaranteed convergence Gradient Ascent algorithms are guaranteed to converge only for infinitesimal steps. The normalization of the mean shift vector ensures that it converges. Large magnitude in low-density regions, refined steps near loc