第4期 张菁，等：图像搜索中人机交互技术的新进展 ·19· 反馈信息量，结合使用反馈信息、多特征高层语义和 time vision for humamrcomputer interaction [M].New 用户模型生成查询表达式.2)基于学习方法生成个 York:Springer-Verlag,2005. 性化文件：根据用户的操作行为、语言描述、专业领 [13]QVARFORDT P,ZHAI Shumin.Conversing with the 域，采用基于学习的方法生成用户模型，构成个性化 user based on eye gaze patterns[A].Conf Humarr Fac- tors in Computing System[C].New York,2005. 文件，如感兴趣搜索词、图像特征或视频特征等，对 [14]TURK M,KOL SCH M.Perceptual interfaces M] 优化搜索结果、过滤非索要信息起到举足轻重的作 Englewood Cliffs:Prentice Hall,2004. 用.3)多模态人机交互：融合多模态和综合使用人们 [15]TURK M,ROBERTSON G.Perceptual interfaces[J ] 的各种感觉器官，使人机交互方式以人为中心、自 Communications of the ACM,2000,43(3):32-34. 然、高效地交互，获得更多的用户语义信息，从而提 [16]SEL KER T.Visual attentive interfaces[J].BT Tech 供拟人化的交互方式。 nology Journal,2004,22(4):146-150. [17]CHEN J BOUMAN C,DAL TON J.Hierarchical brow- 参考文献： sing and search of large image databases [J].IEEE [1]沈兰荪，卓力.小波编码与网络视频传输[M].北京： Trans Image Process,2000,9(3):442-445. 科学出版社，2005. [18]ISHIKAWA Y,SUBRAMAN YA R,FALOUTSOS C. [2]LI Xiaohua,SHEN Lansun.Detecting faces in the wave- MindReader:query databases through multiple exam- let compressed domain [A ]In Proceedings of SPIE ples[A].International Conf on Very Large Data Bases Visual Communications and Image Processing 2005[C]. (VLDB)[C].New York,USA,1998. Beijing,2005. [19]RUI Y,HUANG T.Optimizing learning in image re- [3]LIU Danghui,SHEN Lansun,LAN Kinman,et al.Face trieval [A ]IEEE Conf Computer Vision and Pattern recognition based on illumination restoration[A].In Pro- Recognition[C].South Carolina,USA,2000. ceeding of 2004 International Symposium Multimedia: [20]ZHOU X,HUANG T.Small sample learning during Video and Speech Proceeding[C].Hong Kong,China, multimedia retrieval using BiasMap[A].IEEE Int Conf 2004. Computer Vision and Pattern Recognition[C].Hawaii, [4]FASEL B,LUETTIN J.Automatic facial expression a USA,2001. nalysis:a survey[J ]Pattern Recognition,2003,36(1): [21]CHEN Y,ZHOU X,HUANG T.Oneclass SVM for 259.275 learning in image retrieval [A].International Conf on [5]OUDEVER P.The production and recognition of emo- Image Processing[C].Thessaloniki,2001. tions in speech:features and algorithms[J].Int J of Hu [22]WU Y,TIAN Q,HUANG T S.Discriminant EM al- mamComputer Studies,2003,59(1-2):157-183. gorithm with application to image retrieval [A].IEEE [6]MARCEL S.Gestures for multi-modal interfaces:a re- Conf Computer Vision and Pattern Recognition [C]. view[R].Technical Report IDIAP-RR 02-34,2002. South Carolina,USA,2000 [7]HU Weiming,TAN Tieniu,WANG Liang,et al.A [23]MACARTUR S,BRODLEY C,SHYU C.Relevance survey on visual surveillance of object motion and behav- feedback decision trees in content-based image retrieval iors [J].IEEE Trans on Systems,Man,and Cybernet- [A ]IEEE Workshop CBAIVL C].South Carolina, ics,2004,34(8):3. USA,2000. 8]DUCHOWSKI A.A breadth-first survey of eye tracking (24]TIEU K,VIOLA P.Image retrieval [A].IEEE Conf applications[J].Behavior Research Methods, Instru- Computer Vision and Pattern Recognition C].South ments,and Computer,2002,34(4):455-470. Carolina.USA.2000. [9]PORTA M.Visionbased user interfaces:methods and [25]TONG S,CHANG E.Support vector machine active applications[J].Int J Humarrcomputer Studies,2002, learning for image retrieval[A].ACM Multimedia[C]. 57(1):27.73. Ottawa,Canada,2001. [10]DURIC Z,GRA Y W,HEISHMAN R,et al.Integra- [26]TONG S,KOLL ER D.Support vector machine active ting perceptual and cognitive modeling for adaptive and learning with applications to text classification[A ]In- intelligent humamcomputer interaction [J].Proceedings ternational Conf on Machine Learning [C].Stanford, of the IEEE,2002,90(7):1272.1289 USA.2000. [11]OVIATT S,DARRELL T,FL ICKN ER M.Multimo- [27]VASCONCELOS N,LIPPMAN A.Bayesian relevance dal interfaces that flex,adapt,and persist [J ]Commu- feedback for content-based image retrieval [A].IEEE nications of the ACM,2004,47(1):30-75. Workshop CBAIVL[C].South Carolina,USA,2000. [12]KISACANIN B,PAVLOVIC V,HUANG T.Real- [28]WONG S,ZIARKO W,WONG P.Generalized vector 1994-2008 China Academic Journal Electronic Publishing House.All rights reserved.http://www.cnki.net反馈信息量 ,结合使用反馈信息、多特征高层语义和 用户模型生成查询表达式. 2) 基于学习方法生成个 性化文件 :根据用户的操作行为、语言描述、专业领 域 ,采用基于学习的方法生成用户模型 ,构成个性化 文件 ,如感兴趣搜索词、图像特征或视频特征等 ,对 优化搜索结果、过滤非索要信息起到举足轻重的作 用. 3) 多模态人机交互 :融合多模态和综合使用人们 的各种感觉器官 ,使人机交互方式以人为中心、自 然、高效地交互 ,获得更多的用户语义信息 ,从而提 供拟人化的交互方式. 参考文献 : [1 ]沈兰荪 ,卓 力. 小波编码与网络视频传输[ M ]. 北京 : 科学出版社 ,2005. [2 ]L I Xiaohua , SHEN Lansun. Detecting faces in the wave2 let compressed domain [ A ]. In Proceedings of SPIE: Visual Communications and Image Processing 2005 [ C]. Beijing ,2005. [ 3 ]L IU Danghui , SHEN Lansun , LAN Kinman , et al. Face recognition based on illumination restoration[A ]. In Pro2 ceeding of 2004 International Symposium Multimedia : Video and Speech Proceeding [ C ]. Hong Kong , China , 2004. [4 ] FASEL B , LU ETTIN J. Automatic facial expression a2 nalysis: a survey[J ]. Pattern Recognition , 2003 , 36 (1) : 259 - 275. [5 ]OUDEVER P. The production and recognition of emo2 tions in speech : features and algorithms[J ]. Int J of Hu2 man2Computer Studies , 2003 , 59 (1 - 2) :157 - 183. [6 ] MARCEL S. Gestures for multi2modal interfaces: a re2 view[ R]. Technical Report IDIAP2RR 02 - 34 ,2002. [7 ] HU Weiming , TAN Tieniu , WAN G Liang , et al. A survey on visual surveillance of object motion and behav2 iors [J ]. IEEE Trans on Systems , Man , and Cybernet2 ics , 2004 , 34 (8) :3. [8 ]DUCHOWSKI A. A breadth2first survey of eye tracking applications [J ]. Behavior Research Methods , Instru2 ments , and Computer , 2002 , 34 (4) :455 - 470. [9 ] PORTA M. Vision2based user interfaces: methods and applications[J ]. Int J Human2computer Studies , 2002 , 57 (1) :27 - 73. [10 ]DURIC Z , GRA Y W , HEISHMAN R , et al. Integra2 ting perceptual and cognitive modeling for adaptive and intelligent human2computer interaction [J ]. Proceedings of the IEEE , 2002 , 90 (7) :1272 - 1289. [11 ]OVIA TT S , DARRELL T , FL ICKN ER M. Multimo2 dal interfaces that flex , adapt , and persist[J ]. Commu2 nications of the ACM , 2004 , 47 (1) : 30 - 75. [12 ] KISACANIN B , PAVLOVIC V , HUAN G T. Real2 time vision for human2computer interaction [ M ]. New York :Springer2Verlag ,2005. [13 ]QVARFORDT P , ZHAI Shumin. Conversing with the user based on eye2gaze patterns[ A ]. Conf Human2Fac2 tors in Computing System[C]. New York ,2005. [14 ] TUR K M , KOLSCH M. Perceptual interfaces [ M ]. Englewood Cliffs: Prentice Hall , 2004. [15 ] TUR K M , ROBERTSON G. Perceptual interfaces[J ]. Communications of the ACM , 2000 , 43 (3) :32 - 34. [16 ]SEL KER T. Visual attentive interfaces[J ]. BT Tech2 nology Journal , 2004 , 22 (4) :146 - 150. [17 ]CHEN J ,BOUMAN C ,DAL TON J. Hierarchical brow2 sing and search of large image databases [J ]. IEEE Trans Image Process , 2000 , 9 (3) : 442 - 445. [ 18 ]ISHIKAWA Y , SUBRAMAN YA R , FALOU TSOS C. MindReader : query databases through multiple exam2 ples[ A ]. International Conf on Very Large Data Bases (VLDB) [C]. New York , USA , 1998. [19 ]RU I Y , HUAN G T. Optimizing learning in image re2 trieval [ A ]. IEEE Conf Computer Vision and Pattern Recognition[C]. South Carolina , USA , 2000. [20 ] ZHOU X , HUAN G T. Small sample learning during multimedia retrieval using BiasMap [A ]. IEEE Int Conf Computer Vision and Pattern Recognition[ C]. Hawaii , USA , 2001. [21 ]CHEN Y , ZHOU X , HUAN G T. One2class SVM for learning in image retrieval [ A ]. International Conf on Image Processing[C]. Thessaloniki , 2001. [22 ]WU Y, TIAN Q , HUAN G T S. Discriminant EM al2 gorithm with application to image retrieval [ A ]. IEEE Conf Computer Vision and Pattern Recognition [ C ]. South Carolina , USA ,2000. [23 ] MACARTUR S , BRODL EY C , SH YU C. Relevance feedback decision trees in content2based image retrieval [ A ]. IEEE Workshop CBAIVL [ C ]. South Carolina , USA , 2000. [24 ] TIEU K , VIOLA P. Image retrieval [ A ]. IEEE Conf Computer Vision and Pattern Recognition [ C ]. South Carolina , USA , 2000. [25 ] TON G S , CHAN G E. Support vector machine active learning for image retrieval[ A ]. ACM Multimedia [ C]. Ottawa , Canada , 2001. [26 ] TON G S , KOLL ER D. Support vector machine active learning with applications to text classification[ A ]. In2 ternational Conf on Machine Learning [ C ]. Stanford , USA , 2000. [27 ]VASCONCELOS N , L IPPMAN A. Bayesian relevance feedback for content2based image retrieval [ A ]. IEEE Workshop CBAIVL [C]. South Carolina , USA , 2000. [28 ]WON G S , ZIAR KO W , WON G P. Generalized vector 第 4 期 张 菁 ,等 :图像搜索中人机交互技术的新进展 · 91 · © 1994-2008 China Academic Journal Electronic Publishing House. All rights reserved. http://www.cnki.net