| 本文已被：浏览 4637次 下载 5744次 |
码上扫一扫！ |
|
|
| 基于遗传算法重采样的人脸样本扩张 |
|
陈杰1, 陈熙霖1,2, 高文1,2
|
|
1.哈尔滨工业大学,计算机学院,黑龙江,哈尔滨,150001;2.中国科学院,计算技术研究所,ICT-ISVISION 面像识别联合实验室,北京,100080
|
|
| 摘要: |
| 无论是对人脸检测还是人脸识别来说,训练或测试一个分类器都要进行数据的收集,目前所有基于统计学习的方法都存在这个问题.提出了一种针对已有的人脸样本通过采用遗传算法进行重采样来扩张样本的算法.其基本思想是,基于人脸样本由有限的部件构成,而且遗传算法可以用于模拟自然界中的遗传过程.这种模拟可以涵盖人脸的一些变化,比如不同的光照、姿态、饰物、图片质量等.为了证明该算法所生成样本的推广能力,将这些生成的样本用于训练一个基于AdaBoost的人脸检测器,并且将它在MIT+CMU的正面人脸测试库上进行了测试.实验结果表明,通过这种方法来收集数据可以有效地提高数据收集的速度和效率. |
| 关键词: 人脸检测 遗传算法 SnoW(sparse network of winnow) AdaBoost |
| DOI： |
| 分类号: |
| 基金项目:Supposed by the National Natural Science Foundation of China under Grant No.60332010(国家自然科学基金);the National High-Tech Research and Development Plan of China under Grant Nos.2001AA114190,2002AA118010,2003AA142140(国家高技术研究发展计划(863));the"100 Talents Program"of the Chinese Academy of Sciencesunder Grant No.20056106(中国科学院"百人计划");the Stake of the ISVISION Technologies Co.,Ltd.of China under Grant No.20009040(银晨智能识别科技有限公司资助) |
|
| Face Samples Expanding Based on the GA Re-Sampling |
|
CHEN Jie,CHEN Xi-Lin,GAO Wen
|
| Abstract: |
| Data collection for both training and testing a classifier is a tedious but essential step towards face detection and recognition. All of the statistical methods suffer from this problem. In this paper, a genetic algorithm (GA) based method to swell face database through re-sampling from existing faces is presented. The basic idea is that a face is composed of a limited components set, and the GA can simulate the procedure of heredity. This simulation can also cover the variations of faces in different lighting conditions, poses, accessories, and quality conditions. To verify the generalization capability of the proposed method, the expanded database is used to train an AdaBoost-based face detector and test it on the MIT+CMU frontal face test set. The experimental results show that the data collection can be speeded up efficiently by the proposed methods. |
| Key words: face detection genetic algorithm SnoW (sparse network of winnow) AdaBoost |
