Optimal 2D Model Matching Using a Messy Genetic Algorithm利用遗传算法优化二维模型匹配.pptVIP

下载本文档

2
0
约1.08万字
约 26页
2017-03-08 发布于上海
举报
版权申诉

Optimal 2D Model Matching Using a Messy Genetic Algorithm利用遗传算法优化二维模型匹配.ppt

1、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。。
2、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。
3、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)。
4、该文档为VIP文档，如果想要下载，成为VIP会员后，下载免费。
5、成为VIP后，下载本文档将扣除1次下载权益。下载后，不支持退款、换文档。如有疑问请联系我们。
6、成为VIP后，您将拥有八大权益，权益包括：VIP文档下载权益、阅读免打扰、文档格式转换、高级专利检索、专属身份标志、高级客服、多端互通、版权登记。
7、VIP文档为合作方或网友上传，每下载1次，网站将根据用户上传文档的质量评分、类型等，对文档贡献者给予高额补贴、流量扶持。如果你也想贡献VIP文档。上传文档

Optimal 2D Model Matching Using a Messy Genetic Algorithm利用遗传算法优化二维模型匹配

A Statistical Assessment of Subject Factors in the PCA Recognition of Human Subjects Geof Givens*, J. Ross Beveridge, Bruce A. Draper David BolmeComputer Science, Colorado State University*Statistics, Colorado State University CVPR Workshop: Statistical Analysis in Computer Vision June 22, 2003 Some Human Subjects Are Harder to Recognize than Others. Why? Better Formulate the Question Choose an algorithm Something standard and simple. PCA using our publicly released version. Choose a distance measure: Yambor Angle is good and similar to Moon-Phillips FERET. Much better than standards such as L2. Define Collect Covariates: Few covariates were collected initially with FERET images. One person at CSU, David Bolme, scored all images. Big questions. What to measure? Covariates in isolation or together? Some Basics:Image Preprocessing Integer to float conversion Converts 256 gray levels to single-floats Geometric Normalization Aligns human chosen eye coordinates Masking Crop with elliptical mask leaving only face visible. Histogram Equalization Histogram equalizes unmasked pixels: 256 levels. Pixel normalization Shift and scale pixel values so mean pixel value is zero and standard deviation over all pixels is one. More Basics:Standard PCA Algorithm. What Covariates? The Final Set of Subject Covariates Collecting the Covariates What to Measure? Recognition Rate on Partitioned Data Measure Recognition Rate for Partitioned Images. Partition images by covariate : e.g. male versus female. Compare recognition rate on different sets. Good Answers a very specific version of the question. Recognition rate is a standard performance measure. Bad Fails to adjust, i.e. control for, other covariates. Recognition rate for a probe set, not a single subject. Hidden dependence on gallery image set. Statistical significance is hard to interpret. What to measure? Intrapersonal Image Pair Distance Measure Distance (Similarity) Between Images For two images of one subje