Human-Robot Interaction based on Haar-like Features and Eigenfaces 学习心得及相关的较好的网站

Human-Robot Interaction based on Haar-like Features and Eigenfaces

 

Adaboost 方法 经过Lienhart et al. [2], [4]. 等人的改进。

论文

[2] R. Lienhart and J. Maydt, “An extended set of haar-like features for

rapid object detection,” in IEEE ICIP 2002, Vol. 1, pp 900-903, 2002.

 [4] A. K. Rainer Lienhart and V. Pisarevsky, “Empirical analysis of detection

cascades of boosted classifiers for rapid object detection,” MRL

Technical Report, Intel Labs, 2002.

 

？？？？The detection technique is based on the idea of the wavelet

template [5] that defines the shape of an object in terms of a

subset of the wavelet coefficients of the image.

 

 

Pre-Learnt User RecognitionSystem  系统 是一个几乎实时的人机交互的，

 

FloatBoost 是AdaBoost的改进 论文:：

FloatBoost Learning and Statistical Face Detection - PAMI 2004, Stan Z. Li

 

 

Haar Eye检测

	Haar eye detect	Message List
Reply | Forward				Message #42802 of 44164 < Prev < Prev | Next > Next >

 

Re: Haar eye detect   Hi, I tried seraching on it and found three different sources. Follow these links:- http://face.urtho.net/ contains XML as well as training dataset http://malic.sourceforge.net/malicpartsdetect_demo.html Contains allmen7pos1000neg7000eye.xml http://www.conv2.com/spboard/board.cgi?id=pds2&action=view&gul=139&page=1&go_cnt/ =0 Use some online webpage traslator to view this page, it contains adaboost txt files (i dun know how to use them). None of these work perfectly. I suggest you request Dr. Davids Face Search DLL (available for non-commercial purpose) http://mimban.smb.man.ac.uk/downloads/face_demo.php OR Use AAM-API to create yourself a eye detector http://www2.imm.dtu.dk/~aam/aamapi/ There are some web demos available such as face beta http://www.betaface.com/ (provides a binary sdk too) some commerical eye detectors perform good too such as verilook SDK 2.0... version 3.0 doesnt work as well. the demo sdk comes with a DLL to use Please share your expereince on eye detection too :) Zeeshan. D i g i t a l y Y o u r s, (http://www.yahoo.com/preview)

 

 

有 4 types of Boosting methods are supported such as Discrete AdaBoost, Real AdaBoost Gentle AdaBoost etc.

Face Detection 介绍了很多论文和相关的图像数据库，并且有代码：

http://vision.ai.uiuc.edu/mhyang/face-detection-survey.html

inflomatik  也是有关人脸的

http://inflomatik.com/