о뷢չ
JOURNAL OF COMPUTER RESEARCH AND DEVELOPMENT
1999Vol.36No.6P.641-652



MPEG-4״о
ġ
ժҪżͨżķչýһµʱMPEG-4ʱ׼ڶı롢ģ͵ıȵڶ뼼ڳ.ȴMPEG-4ֵ֧ĸƵص㡢ͲͬӦõĿ/3µı׼Ҫݺ״.µı׼ʵӦûҪṩı׼֮׹ߺо.УͼƵķָȫͼɡļ١2Dģ͵ĽԼ3Dķͺϳɵоͽչ.
ؼʣý弼MPEG-4ڶı룬ģ͵ı룬ͼָͼķϳɣӾ
ţTP391
RESEARCHES AND DEVELOPMENTS OF MPEG-4 CODING
GAO WenWU Feng
(Department of Computer Science, Harbin Institute of Technology, Harbin 150001)
GAO Wen
(Institute of Computing Technology, Chinese Academy of Science, Beijing 100080)
AbstractWith the developments of computers and communications, the multimedia coding comes into a new era. The MPEG-4 standard indicates the mature of the second generation coding technologies, such as object-based coding and model-based coding. In the paper here, all video objects and their characters, scene description, and profile/level in MPEG-4 are introduced. With the demands of the new standard, many non-normalization tools and relevant technologies are provided for the MPEG-4 implementation. In the following content, some key technologies, such as image and video segmentation, panorama generation, facial detection and tracking, 2D mesh generation and tracking, and analysis and synthesis of 3D facial model, are also discussed. 
Key wordsmultimedia technology, MPEG-4, object-based coding, model-based coding, video segmentation, image analysis and synthesis, computer vision
1
ISOMPEG-4׼ĵһ19991ʽ1ݣ׼ĵڶҲ199912¹2.֮MPEGǰͼѹ׼ԣMPEG-4ΪýѹṩһΪƽ̨ඨһָʽͿܣǾ㷨34.MPEG-4ĳϣһɵͨзǿϵͳмµ㷨Ϊü롢ṩ˸ķ56.ԽָĶý弼ڱУѹһЩߡ㷨ͼͺϳɡӾͼΡʵϳɵȼ.MPEG-4֮Ϊڵ绰ϴƵƵݣһͱ˶ͼѹ׼о룬ݺͽҪѾԶԶ˼.
˻ڶ(object-based)ı7,9ݡģ(model-based)1012ݵıȵڶ뼼MPEG-4׼ҪνĶһܹʺͲݵʵ壬ĻֿԸص˶״ģͺ͸߲Ϊ.ֱһֻݵѹʽǰѹ㷨ֻȥ֡ں֡࣬MPEG-4ҪͼƵķ.罫ͼָΪ˶;ֹıƽ棬зֱ.ѹȽϸߡʧȽϴİ취б룬˶ѹȽϵ͡ʧСİ취ѹЧʺͽͼõϺõƽ.MPEG-4ĻԪǶЩǵͶƵ2D3DߵĿĿƵͼΡֵ.ڶıݵѹȣʵݵĽԹܣݵĶýݴȡϷýͥ༭Ϲ͵̵ꡢԶ̼ءҽƺͽѧ.ýѹķչƽǻݵѹʵϢĸ߽׶ΣϢʽ.˵ϢǻźŵģǻһȽϳġֱܹӽмʹķʽ.ýҲݵĴѹ洢ʹ䣬ʹ͸׺Ҳ򵥷.MPEG-4Ӧַչƶƶһýѹ׼.
ڶķּMPEG-4ṩһµĹܣͬʱMEPG-2׼еͼּܣּҪڻխƵͨšƵͶýݿԤȷ.MPEG-4ṩֻķּߣʱּͿּʱּǽԭƵе֡ʣּǽԭƵеķֱ.ÿּУƵжԷΪ㣺ǿ㣬ṩƵеĻϢǿṩƵиߵķֱʺϸڣԵͽ룬ǿһͽ.MPEG-4Ҳ֧ʱͿĻϷּ.
ƶͨŵѸٷչͨƵƵϢԽԽҪˣҪṩ״ͨŻʵְȫĵʱʹ.MPEG-4ı³Ժ;ܣ3ﵽĿģͬ(resynchronizationݻָ(data recovery)( error concealment).ͬڼ⵽ʱڽ佨ͬ㣬ǰһͬ½ͬǷݣݻָͨ׼ṩһֿ䳤뼼ָ֮ͬݣͨռԺƵǰ֡ԶԴͼ.
2MPEG-4Ҫ
ͼ1MPEG-4ĸƵƵɳӣһڽεԼӦɵµƵ2DıSprite3DӺ3Dĵǵ.MPEG-4вЩΪزµĳҲʹ߲ЩѸֶһ˳λ÷һضĳϵУı3D ӽǣʽ򿴳еĳһͨʹĳһ.ǾͷֱMPEG-4ֵ֧ĸֶ롢Ӧÿ/3۸ñ׼ں.


ͼ1MPEG-4Ӧʾͼ
2.1MPEG-4Ҫͼص
MPEG-4ָ֧ƵƵı1415ݣЩȿȻģҲǺϳɵģȿƽģҲ.MPEG-4׼ṹĿԣʵӦõҪԺµĶ.
1ȻƵ
ȻƵǴȻͼƵзָĳһƵͼ2(a)Newsͼеĵһ֡еŮԱΪǰƵ󣬴ʱǹľΣҲһͨڣбȽϸӵ״ͼֻƵһ.ΪƵ״MPEG-4רŶͼСͬĦƽ(alpha plane), ֵƽֱ0255ʾһͼصĹ.ͼ2(b)ͼ2(a)Ķֵƽ棬ɫʾڵڵǰ󣬺ɫʾزڵǰ.ֵƽıûڿ˶ͻڿصĶϵķƵϢɫϢ˶״ӦSADCT(shape adaptive DCT).ƽĶ廹չ0255Ҷȣȿʹǰӵʱڱ߽̫ԡ̫Ӳһ¡ģֿƵʵְ͸͵ƵӾЧ.


ͼ2
2 Sprite
SpriteԱصģ1517.ͨ±ûκ˶ͼеı仯˶ǰ˶ɵģͨͼǶеıͼƴӳһȫıͼͼͽSpriteͼ.SpriteͼֻҪ봫һβ洢ڽˣͼֻҪڱ˶ͿԴSpriteϻָеͼıƵƵ绰ȳ̶ӦУȫͼȴȥڽ.ͼ3SpriteӣϽǱSpriteͼϽǰƵ󣬴ͼǿԿΪʵֱҪǰǰͱͼҪܺܺõķָ;Ҫ޺ۼĴһƵһЩͼƴӳSpriteͼ.


ͼ 3Sprite ʾͼ
3 2D
2Dһ2DͼƽֳεСƬƬƬ֮ȲصҲûз϶ƬĶĽڵ.MPEG-4ĿǰֻɵĶ̬ģͣνĶ̬ģͲָ״˽ṹĽڵһʱڵ˶Ϣ.ͼ4ʼ2Dģͣͼ4(a)һģͣͼ4(b)ǻݵģ.ͼеı仯ĽڵҲƶɽڵɵƬҲᷢα䣬ͨ任˫Բֵǰһ֡ͼӳõǰ֡عͼ.2Dģ͵ıָĽڵλ()˶ǵıǲڽڵԤͱ䳤룬˽ṹҪ룬Delaunay㷨Ψһȷ.


ͼ4
4 3D
3D3D߿ģ״Ϳα仯ȸ沿MPEG-4ײ״˶沿FDP(facial define parameter)沿FAP(facial animation parameter).ͼ5(a)MPEG-4һĬϵ3Dģͣͼ5(b)ģͽɫģFAP¾ĸ˶FAPΪ10˵沿68ֻ˶7.Ϊ3Dģڽ˵õһʵеĺϳͼ񣬿ض˵FDPFDPꡢꡢıȡ沿Ͷض˵.FAPֱ뷽֡ıͻDCTı.֡ıǲϣDCTıǲDCT任ͱ䳤.DCTı뷽ܵõõѹЧʣ˵ҲһЩ.


ͼ5
3Dݽ׼İ汾2.3DһҲ嶨BDP(body define parameter)嶯BAP(body animation parameter)״ơ˶.MPEG-4ͬһĬϵģͣͼ6ģ͵ļͼ29Ҫؽڣÿؽڶ3ɶDOF(degree of freedom), BAPΪ19鹲175.ΪܺϳһضˣBDPƤ̬ơȵȲ.ģ͸ؽڵ˶ûκμƣͨBAPܿģģʵ˵˶.MPEG-4 3DĲֱ׼VRMLH-Animһµ.


ͼ63Dģ͵ļͼ
5̬
ǰ۵2D3DԼ3DıУΪڽ˵õʵеĺϳɶ󣬳ҪûͼɫѶʵʵӳ䵽ģ.ЩϢֻҪڵһ֡һЩؼ֡봫䣬˰ЩҪ봫Ϣ̬.
ͼ7ʾȫӽϢ,ͼѾƥͼ5(a)3DģͣƽϢͼǸͬӽǵͼ񣬰ʽ1

ͼ7ͼȫӽϢͼ
1
3DռĵP(x,y,z)ӳ䵽2DϵĶӦPs(,h).άȣhռ䱻3Dģ͵Ľڵ㻮ΪεƬӳЩƬΪλÿһӳ䵽ıģ͵ĶӦ.̬ıС任ϵķ̬Ҳ״.
6 3D
3DģڼͼΡCADԪͿӻȷŹ㷺ӦãMPEG-4ڱ׼İ汾2ṩһЧı3Dģ͵Ĺߣ3Dģ͵ļ״˽ṹȣάıͨõ3Dı18ݣͼ8(a)һ3DxyУzǸ߶ϢҪֻи߶ϢΪ˲ùʽ2zйһ32λ.3DģͿһά洢ԭеͼ뷽Ϳڹ3D .
(2)
ͼ8(b)һͨ3D3DıҪͼʾĵ㡢ߡԼϢı.Ŀǰйصı뼼MPEG-4 SNHC֤ģϵͳно.


ͼ8
7Ƶ
MPEG-4ƵԷΪࣺȻƵͺϳƵ.ȻƵ2kbit/s64kbit/sʷΧڱ룬ΪʷΧڶõƵMPEG-43ֱ롢뱾ԤʱƵ룬ȻƵıַָּ֧빦ܺʹָ.ϳƵṹƵ(structured audio)ת(text to speech)ṹƵһMIDIԣܱMIDIǿתǰϢϳΪϢⲿӦʱͨ3DеĴϳɽʹ.
2.2
MPEG-4ԲͬĶöı뷽ԵõϸߵѹȣһṩݵĽΪ˽ЩڽһĶý峡ҪһЩӵϢЩʲôʱʲôλá֮ĹϵԼĳ꣬ЩϢͽг(scene description).ͨƳʽBIFS(binary format for scenes)ɵģϢһ˶Գб༭ʱҪ.Ϊӳı༭ͽܣͼ9ʾ״ṹÿҳڵ㶼һڵ㣬ӦһκһҳڵĸڵǻϽڵ㣬ϽڵҪڳı༭.ʵʵӦֽṹǾ̬ģܹӡɾ͸ıڵ.


ͼ9ṹͼ
MPEG-4ûʽֶֽʽԷΪࣺû˽ͷ˽.û˽Ǹı䳡ڵԣʹĳɼʧ̸ͨıλû3Dӵ͸ııͳߴȣЩڽɣҪı.˽ͨûڽ˵ĲҪӦķӳֽҪͨ.ǽVRMLĶֵʽBIFSϣϢԲοISO/IEC14772-1.
2.3MPEG-4Ŀ(profile)ͼ(level)
MPEG-4ṩ˴ġḻƵƵı빤ߣָܹӦҪ.ĳһضӦãֻһϵͳƵƵı빤߱ãܾضӦȷҪõı빤ߣMPEG-4ṩĹ߼һӼ.ÿһһƼĸӶ.ͼ10ʾMPEG-44ܣƵܡƵܡͼοܺͳ.


ͼ10MPEG-4ͼ
3MPEG-4صо
ȻMPEG-4׼Ҫݡ빤ԼʽѾȷMPEG-4µı뼼ǽͼϳɡͼѧʵͼӾȻϣЩµı뼼ҪʹûҪϴĹߺоڶıҪͼƵָߡSpriteҪȫͼ񡢻ģ͵ıҪģͺ͸ģ͡3DıҪͺϳɹߵ.Щ͹߲ڱ׼֮УΪ׼ĿŲо潫Що.
3.1ͼƵķָ
ͼָǼӾѵ֮һȻָûһķ1921.ͼָѵڷָϢֲͳ״˶ȣҪ÷ָĽȷȡһѭ֮.ȻͼƵָУûһ̶״ͽṹϢӾоҶϢԶͼзָ߲ε״˶ȼϢҲԸȷķָҪ߲ϸε֪ʶ.ͼķָʹõķָϢǰͼķָΪķָ˶ķָͻʱյķָ3.
ͼƵָõļСٸȡɫʺ˶ȸѡһʵֵǰӱзָķָ22,23.ھ㷨԰ҪָĶͼָK24,25ݣַָҪ֪ȷķõȿһҲǸ߽׵.ͼͳԣøͳƵķзָMAP(maximum a posteriori probability);GibbsֲɷMRF(Markov random field)ȣ2628.ѧ̬ѧеˮ㷨һЧͼָߣ2930ݣˮ㷨ָͼķǻķͼнϺõ³.
ܶͼķָоһЩضͼԶָܵõϺõЧ򵥱˵ͷͼȣӻ»ͼԶָЧܲ.ΪʵMPEG-4ĻڶıָǱزٵģҪһЧķָߣùܹĶԶָĽ޸ĺ͸Զָṩ߲ϢΪ߷ָЧʣڶĸټ3132Ҳ.
3.2ȫͼ
ͼǶԶĳһ3DȫͼһµоͳӦǽңͼͼһĵͼ33ݣŸüķչӦõķΧҲԽԽ3436.µMPEG-4׼о෽ҪֱӻõͼǶSprite롢3D ıлͼɫԼݵƵ.ȫͼͨļõȫֱӰһ㵽һγƬϣôӽǵȫλ㣬ЩҪ豸ʵõļǽͨͼһƵǶȫͼ3739.
ͼλùϵһͼе˶仯ҪһϵǷͼĻ.ϵȿ3Dֿ2Dͼƽ꣬ݲͬϵõ˶ģҲͬ.2Dͼƽͨùʽ3ͶӰģͼ֮ĹϵģͲ3Dڵ˶ͼı仯Ϊ˶.ͼǶҲùʽ453DתϵĽѾ֪Ϊ˽һȫͼֻҪƸͼճǶȺʹֱƶ.

3D˶Ҳֱ3DռΪԭ㽨3Dĳ꣬ռϵĵPͼƽϵĶӦXĹϵʽ6TλƱ仯VĽ仯R3Dת.ģ֡ͼ֮˶ϵͨС׼˶ģ͵ĲЩͼ˶ǵʵλǶȫͼ.
X=TVRP(6)


MPEG-4ҪṩͼǶ׼ȷٵһͼеȫͼȫͼַܳ϶ӰSpriteܼ½.⻹ҪͼǶϵȥ߶̬Sprite뼼.
3.3ļ
Ƶ绰ƵͷеıУļǻģͱĻMPEG-42D3Dı붼Ҫṩʵʱ͸ٹΪģ͵ƥṩ.ΪڸӱԼպȷؼ͸༼Ӧõⷽо.
̶ģ巨һ򼸸οģ壬Ȼοģ֮ƥȣֵͨжͼĳһǷַȽϼ򵥣ʵӦֻΪԤֶΣ40.ģ4142̶ģ巨ͬģаһЩǹ̶Ԫ.һַֹߺԱеĳЩǹ̶۾Ӻ촽.һַϵͳԶӦߺ棬Թɱģ.򷨣4344ǴͼĽṹмģΪȻеһضӳͼϣһĽṹֲ.ͨͼǷЩЩĳ̶ȷͼǷ.ѧϰ4546ǽΪӷʶģʽʶ⣬ͨͷѧϰձõķ˹.ɫϢķɫڲɫռеķֲԽϼУ͸.÷һɫģɫһйģͶͼͲȻλ͸.Eigenfaceʵɷַ47ݺK-L任.ף48(eigenface)ĸΪֱͬͼɵĿռһӿռ䣬ĿͼӿռͶӰ֮ľΪ.
Ҫ˶ (motion-based ) ķͻģ (model-based )ķ49,50.һûģ͵ķ˶ģϵķ5153.MITPentlandTonyһֻ3Dģ͵ĸٷ.ͨ˹ʱ˲ȡ˶Ŀÿ˲ֱĿ꣬Ȼ3Dģƥ49.,ɫɫȿռзֲΪ,ɫϢɫģҲһּЧķ52.
ΪڸӱԼպȷؼ͸ͨҪһж༶ṹģͣ54, ֱȡͬʵĹ, ﵽĿ.
3.42Dģ͵Ľ
MPEG-42Dı涨˱ڵλú˶ʸķ͸ʽǷǹĽڵλѡܸȷعƽڵ˶иߵرЧʣԼθı仯ΪһԵдһо.
555758򵥵ѡ񣬰һĹѡȡģ͵ĽڵȻܸЧرڵλãڵλѡȡԼӲܹӳͼĽṹͱ߽磬һƬлڲͬ˶ͲͬԣӰ쵽˶Ƶ׼ȷԺӳЧ.Ӧģͣ5661ڹģ͵ĻͨѺͺϲѲҪƬΪСƬģͿһ̶Ϸӳͼݺͽṹ.ͼݵģҪǲͼģͣõģܺܺõķӳͼı߽.಻ͬͼɻݵģͣ6063.2DͼıԲͨõģȥƥ䣬ģͰ֪ʶиߵıЧʣ64.
Ƭ˶仯ͨ6ģģ͵˶仯˽ṹֲ䣬ڹڵ˶ʱڽڵλøıӰ˶ƵĽģ͵˶ҪȫŽ.Brusewitz57TBM(triangle-based motion compensation)Ȳÿƥ似ڽСķΧڹƸĴµ˶Ϊ˾ȷж˶TBMûݶȵ˶Ʒͬʱж˶.ȻǰĿƥ似ܸݶṩõĳʼֵTBMļ൱.Nakaya58,59ƥTBMݹ˽ṹص㣬ÿ(߽)6ɵεģõ˶εڹƣΪŵ˶Ʋ˵㷨.
ܴ֮Ųͬļģ͵˶ƺ͸٣ЩǼ̫޷ʵʱԵҪ󣬾ǲܱ֤ȫŽ⣬ҪĸŻķ̽Ч㷨.
3.53Dģ͵ķͺϳ
3DģͶı漰Ҫĸ.ʵе3Dģ⼼;ʵе˼ڿӵ绰ӻ顢⻷͵ӰؼҪӦ.Ϊܵõض˵FDPBDPҪкʵĹȡЩΪڽģʵ˵ı顢˶ơߵ˶Ƶ˶ҲҪȡӦFAPBAPʵ˵ʹʵеҪأҲҪṩͼɫ.
3Dģ͵ıУΪ׼ȷظ3D˶߱Чʺ͵õĽͼģʵ˵ͼƥûҪ.ͬ˵ͷС״Լ۾ͱӵλöһĲҪض˵ͼͨõ3Dģ.Aizawa6566ݵȲ3Dӳ任ƥһض˵ͼͨ˻ʽͼĶ㡢̫Ѩüĵ4㣬ȻЩ3Dģ͵Ľڵλʹ֮ƥͼ.Kaneko67ݵһʽ3Dģ͵ķַֻǱ궨7.Huang68ݵͼʱݶȹĳȺͿģ͵ĳ߶ȣȻٱ궨ϸģ.˶Զ3Dģ͵ķҲо6970ݣԶ3Dģ漰ļ١ʶԼƥȼ㷨˵Ƚϸ.ЧĽʽ3Dģƥ乤ǱȽʵʵһѡ7172.
3Dģ͵ĸҲȹģ͵˶ٵصĸڵ˶.˶ǻһ˶ģЩ˶ķǰȫͼ÷ͬ.
3DģƤɫܿӽˣģ͵ıͨõɺʵͼ.ƤожƤ7374ݣIshii75ݵ1993һ˵ƤģͣģɣƤļģͺƤı淴ģͣƤļģͷӳƤƵƤı淴ģͷӳʵƤĹ.ƤϢķȻڱӦвôϢɵȱʵ.Ϊ˵õʵеĺϳͼͨʵͼģͽɫ.ʵϢȿͬǶȵĶͼҲͼ7ʾȫͼȫܸϢΪ˵õȫͼҪӶͼǶ.
4ᡡ
MPEG-4µı׼Ӧý漰оͷչͼѧӾģʽӿڡVRȣͬʱΪָĶýϢṩͳһѹʽ.ʵҳе˹ڶ๦ܸ֪ĶҪ⣬顢ȶϢķʶо˲ٹ顢ʹĺϳɷҲȡ˲ٳɹ.MPEG-4׼ĹͼıȶýϢͨͳһĹ淶ͽͬʱϢķͺϳɵоҲʹMPEG-4׼ʵû.
ĿõȻѧ(Ŀ69789301)ҡ߿Ƽ(Ŀ863-306-03-01-1).
߼飺ģУ19564ڣʿʦйѧԺ㼼оҡƻܼר鳤(ϯר)оԺԺҪоΪ˹ܺͶý弼ڼӾģʽʶͼýѹ.
㣬У19697ʿоҪоΪ˹ӦúͶý弼ڼӾģʽʶͼýѹģʽӿԼʵ.
ߵλģҵѧѧϵ150001
㣨ҵѧѧϵ150001
ģйѧԺ㼼о100080
οף
1Overview of the MPEG-4 standard. ISO/IEC JTC1/SC29/WG11, N2323. Dublin, Ireland, 1998
2Overview of MPEG-4 functionalities supported in MPEG-4 Version 2. ISO/IEC JTC1/SC29/WG11, N2197. Tokyo, Japan, 1998
3PEG-4 requirements, Version 8. ISO/IEC JTC1/SC29/WG11, N2321. Dublin, Ireland, 1998
4MPEG-4 applications document. ISO/IEC JTC1/SC29/WG11, N2322. Dublin, Ireland, 1998
5Sikora T.The MPEG-4 video standard verification model.IEEE Trans on Circuits and Systems for Video Technology, 1997,7(1):1931
6Pereira F,Alpert T.MPEG-4 video subjective test procedures and results.IEEE Trans on Circuits and Systems for Video Technology,1997,7(1):3251
7Kunt M,Ikonomopouls A,Kocher M.Second generation image coding techniques. Proc of the IEEE, 1985,73(4): 549575
8Kunt M,Benard M,Leonardi R.Recent results in high-compression image coding. IEEE Trans on Circuits and Systems, 1987, 34(11):13061336
9Gerken P.Object-based analysis-synthesis coding of image sequences at very low bit rates. IEEE Trans on Circuits and Systems for Video Technology, 1994, 4(3):228235
10Li H, Lundmark A,Forchheimer R.Image sequence coding at very low bitrates: A review. IEEE Trans on Image Processing,1994,3(5):589609
11Aizawa K,Huang T S.Model-based image coding: Advanced video coding techniques for very low bit-rate applications. Proc of the IEEE, 1995,83(2):259270
12Tekalp A M,Beek P V,Toklu C.Two-dimensional mesh-based visual-object representation for interactive synthetic/natural digital video. Proc of the IEEE,1998,86(6): 10291051
13MPEG-4: Visual committee draft.ISO/IEC JTC1/SC29/WG11, N2202. Tokyo, Japan, 1998
14MPEG-4:Audio committee draft.ISO/IEC JTC1/SC29/WG11, N2203. Tokyo, Japan, 1998
15Irani M,Hsu S,Anadan P.Video Compression using mosaic representations. Signal Processing:Image Communication,1995, 7(46): 529552
16Lee M C,Chen W G.A layered video object coding system using sprite and affine motion model. IEEE Trans on Circuits and Systems for Video Technology, 1997, 7(1):130145
17㣬.̬SpriteоĽ. ѧ1999, 22(3):262268
(Wu Feng, Gao Wen. Research and improvement on on-line Sprite encoding. Chinese Journal of Computers (in Chinese). 1999, 22(3):262268)
18SNHC Verification Model 8.0. ISO/IEC JTC1/SC29/WG11, N2229. Tokyo, Japan, March, 1998
19̵£.Ӿ㷨. ѧ磬1998
(Ma Songde, Zhang Zhengyou. Computer VisionComputing Theory and Algorithm Basics (in Chinese). Beijing: Science Press, 1998)
20Tekalp A M.Ƶ.廪ѧ磬 1997
(Tekalp A M. Digital Video Processing (in Chinese). Beijing: Tsinghua University Press, 1997)
21Dufaux F,Moscheni F.Segmentation-based motion estimation for second generation video coding techniques. In: Torrres L,Kunt M eds, Video Coding: The Second Generation Approach. Boston:Kluwer Academic Publishers, 1996.219263
22Snyder W,Bilbro G.Optimal thresholdingA new approach.Patt Recog Lett,1990,11(5):803810
23Lee S U,Chung S Y,Park R H.A comparative performance study of several global thresholding techniques for segmentation. Comp Vis Graph Image Proc,1990, 52(2):171190
24Pappas T N.An adaptive clustering algorithm for image segment. IEEE Trans on Signal Proc,1992,40(4):901914
25Coleman G B,Andrews H C.Image segmentation by clustering. Proc of the IEEE,1979,67:773785
26Derin H,Elliott H.Modeling and segmentation of noisy and textured images using Gibbs random field. IEEE Trans on Patt Anal and Mach Intel, 1987,9:3955
27Kervrann C, Heizt F.A Markov random field model-based approach to unsupervised texture segmentation using local and global spatial statistics. IEEE Trans on Image Processing, 1995,4(6):856862
28Li S Z.Markov Random Field Models in Computer Vision. New York:Springer-Verlag, 1995
29Beucher S ,Meyer F.The morphological approach to segmentation: The watershed transformation. In: Dougherty E R ed. Mathematical Morphology in Image Processing.New York: Marcel Dekker, 1992
30Vincent L,Soille P.Watersheds in digital spaces: An efficient algorithm based on immersion simulations. IEEE Trans on Patt Anal and Mach Intel,1991, 13(6):583598
31Meyer F,Bouthemy P.Region-based tracking in an image sequence.In: Europ Conf on Computer Vision.Italy: Santa Margherita,1992.476484
32Legters G R,Tzay Y Y.A mathematical model for computer image tracking. IEEE Trans on Patt Anal and Mach Intel,1982,4(6):583594
33Moffitt F H,Mikhail E M.Photogrammetry,New York:Harper &Row,1980
34Irani M,Hsu S,Anandan P.Video compression using mosaic representations. Signal Processing:Image Communication,1995,7: 529552
35Chen S,Williams L.View interpolation for image synthesis. In: SIGGRAPH'93. New Orleans, 1993.279288
36Irani M,Anandan P.Video indexing based on mosaic representations. Proc of the IEEE,1998,86(5):905921
37Szeliski R,Shum Y Y.Creating full view panoramic image mosaics and environment maps. In:SIGGRAPH'97. Los Angeles, 1997.251258
38Kang S B,Weiss R.Characterization of errors in compositing panoramic images. Digital Equipment Corporation. Cambridge Research Lab,Tec Report: 96/2,1996
39Szeliski R.Video mosaics for virtual environments. IEEE Computer Graphics and Applications, 1996, 6(2):2230
40Brunelli R,Poggio T.Face recognition: feature versus templates.IEEE Trans on Patt Anal and Mach Intel,1993,15(10): 10421052
41Yuille A,Hallinan P,Cohen D.Feature extraction from faces using deformable templates.International Journal of Computer Vision,1992,8(2): 99111
42Chow G, Li X.Towards a system for acoustic facial feature detection. Pattern Recognition,1993, 26(12):17391755
43Yang G.Human face detection in a complex background. Pattern Recognition, 1994, 27(1):5363
44Sinha P.Object recognition via image invariants: A case study.In: Investigative Ophthalmology and Visual Science. Florida, 1994.17351740
45Sung K,Poggio T.Example-based learning for view-based human face detection.Tech Report of MIT, AI Memo:1521, 1994
46Rowley H A,Buluja S.Human face detection in visual scenes.Technical Report of CMU: CMU-CS-95-158R,1995
47Pentland A, Moghaddam B. Thad starner view-based and modular eigenspaces for recognition. In: Proc IEEE Computer Soc Cont on Computer Vision & Pattern Recognition. Colorado, USA, 1994. 8491
48Turk M A, Pentland A P. Face recognition using eigenfaces vision and modeling group. Journal of Cognitive Neuroscience, 1991, 3(1):7186
49Mckenna S, Gong S. Tracking faces. In: Proc of the Int'l Conf on Acoustic Face and Gesture Recognition. vermont: killington, 1996. 203206
50Pentland A. Modal matching for correspondence and recognition. IEEE Trans on Patt Anal and Mach Intel. 1995 17:11231129
51Yang J, Pentland A. Real-time tracking of the human body. Tech Rep of CMU: cmu-cs-95-210, 1995
52Gong S, Psarrou A. Tracking and recognition of face sequences. In: Eur Workshop on Combined and Synthetic Image Processing for Broadcast and Video Production. Hamburg, 1994
53Basu S., Essa I. Motion regularizaiotn for model-based head tracking. Tech Rep of MIT: 362, 1996
54. ٵо۲ʿģ. ҵѧ1997
(Liu Mingbao. Researches on facical detecting Ph D dissertation(in Chinese). Harbin Institute of Technology, Harbin, 1997)
55Sullivan G J. Bake R L. Motion compensation for video compression using control grid interpolation In: Proc ICASSP'91. Toronto, Canada, 1991. 27132716
56Huang Chung-Lin, Hsu Chao-Yuen. A new motion compensation method for image sequence coding using hierarchical grid interpolation. IEEE Trans on Circuits and System for Video Technology, 1994, 4(1):4257
57Brusewitz H. Motion compensation with triangles. In: Proc 3rd Int'l Conf on 64 Kbit Coding of Moving Video. Rotterdam, Netherlands, 1990
58Nakaya Y, Harashima H. Motion compensation based on spatial transformations. IEEE Trans on Circuits and System for Video Technology. 1994, 4(3):339356
59Nakaya Y, Harashima H. An iterative motion estimation method using triangular patches for motion compensation. In: SPIE Visual Commun and Image Processing '91. Visual Commun, Vol 1605. Boston, MA, 1991
60Altunbasak Y, Tekalp A M. Closed-form connectivity-preserving solutions for motion compensation using 2-D meshes. IEEE Trans on Image Processing, 1997, 4(3):12551269
61Altunbasan Y. Object-scalable, content-based video representation and motion tracking for visual communications and multimedia Ph D dessertation, University of Rochester, New York, 1996
62Dudon M, Avaro O, Roux C. Triangular active mesh for motion estimation. Signal Processing: Image Communication, 1997, 10: 2141
63Wu F, Gao W, Qu L J. Mesh generation algorithm with morphological tools. In: ICMI. Hong Kong, 1999
64Reinders M J T. Model adaptation for image coding Ph D dessertation, Delft University, Netherlands, 1995
65Aizawa K, Harashima H, Saito T. Model-based analysis-synthesis image coding system for a person's face. In: Signal Proc: Image Comm, 1989, 1(2): 139153
66Aizawa K. Model-based synthesis image coding system. In: Proc Globe Com '87. Tokyo, Japan, 1987. 4549
67Kaneko M, Koike A, Hatori Y. Coding of facial image sequence based on a 3D model of the head and motion detection. Journal of Visual Comm and Image Representation, 1991, 2:3954
68Huang T S, Reddy S C, Aizawa K. Human facial motion modeling, analysis, and synthesis for video compression. SPIE Visual Communications and Image Processing, 1991, 1605:234241
69Huang H, Ouhyong M, Wu J. Automatic feature point extraction on a human face in model based image coding. Opt Eng, 1993, 32: 15711580
70Reinders M T, Sankur B. Transformation of a general 3D facial model to an actual scene face. In: 11th Int'l Conf Pattern Recog. Hague, Netherlands, 1992. 7559
71̽࣬ģŵ. ʵеάضɷ. ڶйͼѧѧ. ֣1998. 142148
(Yan Jie, Gao Wen, Yin Baocai et al. Generation of realistic 3-d specific virtual human face. Chinagraph'98. Guilin, 1998. 142148)
72ݸģ֣̽. 沿Ϊĺϳ. ѧ1998, 21(8): 694703
(Gao Wen, Chen Xilin, Yan Jie, Synthesis of facial behavior for virtual human. Chinese Journal of Computers, 1998, 21(8):694703)
73Miller G. The motion dynamics of snakes and worms. Computer Graphics, 1988, 22(4): 169178
74Kaufman A. Ts1A texture synthesis language. Vusual Computer, 1988, 4(3): 148158
75Ishii T, Yasuda T, Toriwaki J. A generation model for human skin texture. In: Thalman NM, Thalmann D eds, Proc CG International'93:Communicating with Virtual Worlds. Tokyo: Springer-Verlag, 1993. 139150
ոڣ1999-04-08
