一种基于动态贝叶斯网的视觉注意力识别方法 - 人机交互与媒体集成 .pdf

3A Vol.39 No. 3A 2011 3 ACTA ELECTRONICA SINICA Mar. 2011 董力赓,邸慧军, 陶霖密, 徐光 ( ( ), 100084) : ., ., . .. . : ; ; : TP3914 : A : (2011) 3A14007 Dynamic Bayesian Network Based Visual Focus of Attention Recognition DONG Ligeng,DI Huijun,TAO Linmi, U Guangyou ( Dep artment of Computer Science and Technol ogy , Tsinghua N ational L aboratoryf or I nf ormation Science and Technology , Tsinghua Univers ity , Bei ing 100084, China) Abstract : Visual focus of attention recognition is usually based on head pose estimation.However, in a real application, it is difficult to accurately estimate the head pose due to large pose variations,low resolution imagesand varying illuminations.To handle the problem, we propose a dynamic Bayesian network model to infer the visual focus of attention. The head pose is not explicitly computed but measured by a similarity vector which representsthe likelihoodsof multipleface pose clusters.The model encodesthe probabilistic relationsamong multiple foci of attention,multiple user locations and faces captured by multiple cameras.Data are col lected in a prototype ambient kitchen and results show that the model is effective. Key words: visual focus of attention recognition;dynamic Bayesian network;the ambient kitchen , 1 . [ 1] , , (Visual Focus of , . Attention, VFOA) . , , . , , .(eye gaze) , [2] . , . ,