Seamless Integration of Multi-Face Detection and Tracking Meth

CHUR > College Computer Science and Informatics > Department of Computer Science and Information Engineering > Journal Articles > Item 987654321/32180

Please use this identifier to cite or link to this item: http://chur.chu.edu.tw/handle/987654321/32180

Title:	Seamless Integration of Multi-Face Detection and Tracking Meth
Authors:	黃雅軒 huang, yea-shuan
Contributors:	資訊工程學系 Computer Science & Information Engineering
Keywords:	人臉追蹤;多實例學習;遮蔽偵測;遮蔽還原 Face Tracking;Multi-Instance Learning;Occlusion Detection;Occlusion Recovery
Date:	2012
Issue Date:	2014-06-27 01:46:35 (UTC+8)
Abstract:	本論文提出一種與人臉偵測處理緊密結合的人臉追蹤技術。一開始得到人臉區域之後，會先隨機取樣並使用隨機產生的海爾特徵來計算特徵值，接著利用結合多實例學習及Online-AdaBoost等概念的Online-MILBoost演算法來訓練人臉追蹤模型，然後使用訓練出來的追蹤模型來進行人臉的追蹤。為了避免在追蹤時錯誤的持續累積，我們在追蹤到的人臉區域附近，進行人臉偵測處理，當偵測到人臉時，則利用人臉偵測的結果來取代追蹤的資訊，以達到更好的追蹤效果。本論文也對多人臉追蹤時的遮蔽狀況進行改善處理，首先，會利用遮蔽判斷流 This paper presents a face detection process and integration of face tracking technology. Face region to get started after a random sampling will be randomly generated and used to calculate the characteristic features of the value of Haar, and then use wi
Appears in Collections:	[Department of Computer Science and Information Engineering] Journal Articles

Files in This Item:

File	Description	Size	Format
p_e331_0292.pdf		28Kb	Adobe PDF	144	View/Open