ѧ
JOURNAL OF SOFTWARE
1999ꡡ109ڡVol.10No.91999



ͷƵƵо*
ѩ v ʷԪ
ժҪڵӰӵȹƵź,ڵĽų̶һ㶼ھͷлƵϵ,ڶԴƵźŵļУûϣܹڽų̶м.˶ԴƵźŵľͷлƵʽж,ݵõľͷƵȽƵķ,Ƶľͷָ㷨,ԼΪС©/ӰõľͷƵȶͷƽ֡³㷽.»þͷƵȽƵĻ˼뼰㷨ϸ,˶ʵӰƬƵźżʵͷ.
ؼʡƵ,ͷƵ,ͷб,ͷ,ͷƽ֡.
ͼšTP311
Using Shot Frequency in Video Retrieval
BAI Xue-sheng,XU Guang-you,SHI Yuan-chun
(National Key Laboratory of Intelligent Technology and Systems Tsinghua University Beijing 100084)
(Department of Computer Science and Technology Tsinghua University Beijing 100084)
AbstractIn the film, television and many other videos, exciting video sections are usually reflected by their frequent shot changes, and such sections are the most frequently objects in retrieval of this kind of video. In this paper, a method on retrieving videos according to their shot change frequencies is presented. Scene transition analysis has done on real videos, and video segmentation algorithm for both cut and dissolve are developed. To lessen the influence of error and missing shot segmentations, a robust calculation method of shot frequency on shot sequence acquired is proposed. Basic ideas and algorithm modules are given in detail, also with the experimental results on film video and analysis.
Key wordsVideo retrieval, shot frequency, shot cut, chromatic edit, average frame number.
ڶƵźŵĻݵļ,£ûϣܹϢƵƬ.Ȼ,ӵǰý⼼ˮƽ,ͨõġܹԶƵźŽзϵͳڽڻʵ.һĳϻӦ,ĳЩض庬.Щ,רõƵϵͳҲͳΪĿǰƵźŻݼоеһȵ.
Ƶ,ĿǰõĴ²裺(1) ƵָΪжϢƵƬΡͷ2 Ծͷ,ݲͬͷԺԶԾͷ֯,߲ṹ.õõ֯,ûͨѡȤƵƬνм14.
Ȼṩһȫ桢ЧƵֶ,ҪдĽ⹤,Ӧ,ûļҪԲҪƵźŽϸµ.,ڵӰӵƵź˵,ûһϣܹ(򶷡׷𡢶š),ͨͷлƵʵ.༭ĹƵźҲƵ.ص,ԴƵźŽһӳͷлƵʵĶͷƵ,ӶΪûṩһġƵźŸų̶Ƚмķ,ɾ൱Ҫ.
ʵӦ,ҪֱӶԾͷƵʼ㣬ΪͳƾͷĿĵλʱ䲻ȷԹ㷺Ӧ,һֿĶǾͷƽʱ.ʵӦΪͷƽ֡.Ŀǰľͷָȷõȷľͷָ,ڵõľͷ֡©ֵľͷ.,ֱӼƽ֡Ľ.Ϊ˽,߾ͷָ㷨ܵͬʱ,³ľͷƽ֡㷽.ַ,õƽ֡ھͷָеĴнǿĿ,ӶܴӱϷӳƵоͷлƵʣڵĽų̶ȣ,ҲܸõûļҪ.
ĵ1ڶƵɺͱ༭ҪĽ.2ڶԲõľͷָԼ³ͷƽ֡Ʒн.3ڸ˶ʵƵþͷƵȽмʵ.
1 Ƶɼ༭
ƵźŵĲԭ֪,һƵźűɶξͷϴONOFFڼõһƵźţͨƴӱ༭.ھͷڲ,ںƵ֡,仯Сھͷת,Ƶ֡ᷢԵĸı.ת༭ԵĲͬ,ͷתɷΪб䣨cutͽ䣨dissolve.
1.1 ͷб
ͷб򵥵Ƶ༭,ͨͷֱƴӶγ,ͷת䲻κα༭.
ͼ1ʾ,ȷֱΪt1t2ľͷA(t)B(t)оͷб,γһΪt1+t2ľͷC(t).C(t)ĶΪ
1

ͼ1ͷб
ɼ,ھͷбt1֡Էͻ,ط֡仯С.
1.2 ͷ
ͷһƵͷھͷתĳʱڽתΪƵͷı༭ʽ.ʵƵźű༭,ͷһͨԵĹѧӦƵ༭豸.ͷźŹɷʽͼ2ʾ.

ͼ2ͷ
ͼ2,ڳΪt1ľͷA(t)ͳΪt2ľͷB(t)һͷ.ͷA(t)t1ֹ,Ϊte1-t1˥,te1ȫʧ.ͷB(t)źts2ʼ,Ϊte2-ts2,te2ﵽ100%.ɵ¾ͷC(t)Ϊte2+t2,еƵźǾͷA(t)˥źź;ͷB(t)Ľź֮.ֱ徵ͷA(t);ͷB(t)ĳΪte2+t2չͷź
2

3
¾ͷ
c(t)=A*(t)+B*(t),0tte2+t2.4
A(t)B(t)ֱΪźʱ,ͷ˻Ϊ(fade-in)͵(fade-out)ֳƵ༭ʽ.
2 Ƶͷķָ
2.1 ͷбļ
Ծͷбı༭,ȻбӦ֡Եͻ.Ϊ˼ټ,ӿٶ,ǲֱ֡DCͼΪ֡Ե.Ȼ,ھͷڲDCͼ仯С,ھͷб䴦ǰDCͼֱӦǰһͷĩ֡ͺһͷ֡,ͼԽֽϴı仯.DCͼĲ,Ӧһֵĳ.ͨⵥֵ,Ϳȷбλ.
ͼоֽϴȽԵ˶ʱ,DCͼҲֽϴı仯,DCͼгֵַ.,ֱӶDCͼ趨ֵ´.˶,ֵĳҲֲ,ǵֵ.ص,ǲȡDCͼ¶/δֵ.
5
ֵķ.
ڹʽ(5),d1(t)=IDC(t+1)-IDC(t)Ϊ֡DCͼ,wΪڰ뾶,ͷֱΪtΪ,wΪ뾶Ĵڴ֡DCͼеֵʹδֵ.
ʵƵźŵбͼ3ʾ,е1(ӦDiff)ΪƵźDCͼ.3(ӦMax)Ϊõ϶/δֵ.2(ӦAver)Ϊ¶ֵ/ֵ

wڰ뾶Ϊ5.

ͼ3 ͷб
ͼ3,a,b,dֱӦ3ͷб,cӦͼгֵĽϴ˶.пԿ,/δֵпԸܺõб.
2.2 ͷļ
ݵ1.2еģ,ھͷиԺ㶨仯.Դص,Zhang Hong-jiang˼㶨ͼconstant image,CI)Ľⷽ1.ʵʽ̲ȫϸģͣͼڱ仯,ʵӦзָ㷨.ͨ,µĽ˼.
ȣǶw֡DCͼУ
dw(t)=IDC(t+w)-IDC(t).6
ڽ,ھͷڲ,DCͼIDC(t)仯С,Ϊֺ㶨,ͼ4(a)ʾ.,dw(t)Ͼͷ̾ͶӦһ״ķ,ͼ4(b)ʾ.,Ƕw֡DCͼƫ
7
ھͷڲ,Ǿֹʱ,ֲ˶ʹEw(t)һԽϸߵˮƽ.ھͷת,źŵı仯Ա仯Ϊ,¹ֵĳ,ͼ4(c)ʾ.

ͼ4ͷʾͼ
Ϸ,ǲ˶dw(t)Ew(t)ֱȳֵwthresholdķ͹ȵ㷨.ʵ,wΪ10,ֵwthresholdҲΪ10.ͼ5ʾ.
ͼ5(a)(b)ֱΪƵźŵļ,ͼ5(a)еabΪ,ͼ5(b)еcdΪḶ̌ʵǰһĻźһĻĵ룩.

ͼ5 ͷ
2.3 ͷƵȵ³
ͨƵͷָ㷨,Եõһͷ.ǽԾͷоͷлƵʶͷƵȵĹƽ.
ͷƵȵֱӺڵλʱڳֵľͷĿ.ڲͬͷĳʱܴ,ʵӦͳƾͷĵλʱȷ.һ뾵ͷƵӦĶǾͷƽʱ,Դ˾ͷΪĵĸȴھͷʱƽֵ.Ȼ,ͷƵԽ,ͷƽʱԽС.ʵ,ͷĳʱþͷа֡ؼԱʾ.
ֵע,ͷָ㷨ܱ֤ܸȷķָ,ǲɱش©ֺʹ.Ծͷоͷƽ֡,õĽֱܵ©ֺʹֵӰ.,˵ĸ֪Ƕȶ,ڽų̶Ǿͷдͷ֡ƽͳ,ڼҲӦȥ֡֡һЩͷ.
,µľͷƽ֡㷽.
ͷмΪf(n),ֵnͷ֡.ǴڳΪL,SL(n)ΪnΪģΪLĴоͷ֡ɵļ.SL(n)еԪذС˳,ȥСabԪأõ¼ϼΪSa,bL(n),ͷƽ֡ɰ¹ʽ.
8
ͷƽ֡ͨԳΪLȥСabԪؼֵõ.Ȼ,©ɵĳͷʹɵĶ̾ͷԴƽ֡ļӰС,㷨нǿ³.
3 㷨ʵ
ñ㷨,Ʋʵһ׾ͷָ㾵ͷƵԼþͷƵȶƵϵͳ,ôϵͳʵʵӰƬƵ˻ھͷƵȵļ.
ϵͳṩþͷ֡ƵĹ,ͼ6(a)ʾ.,Ϊָõľͷ֡,Ϊ㷨õľͷƽ֡.ھͷƽ֡ļ,ǽڳL趨Ϊ8,ȥСԪظaԪظbΪ2.Ϊڽ,Naver(n)õĴڲnΪ,nΪĴ.ͨѡ˫оͷӦλ,ϵͳͿɴָͷʼƵ.,ûͨѡͷƽ֡онСƽ֡ľͷѡͷлƵƵƬΣһ㶼ӦڽΪŵƬΣ.

(a)(b)
ͼ6 þͷƵȶƵźŽм
Ϊ˶ԾͷƵȵýһ˵,ϵͳṩ˶Ծͷаƽ֡СеĹ,Աûļ.źõĴ֡ͼ6(b)ʾ.ͼʾǶӰƬˡMan In Black()еľͷ֡ŵõĽ,ԣǰĴ֡ӦĻӰƬڽŵĲ.ͼ6(b)д֡˳Ϊ(1) ڲ͹ݵıɱ(2) MIBɵĻң(3) ˹JKǹ֧ԴڰٻĶţ(4) ˹JMIḄ(5) ˹J˵׷.г(4),ڽΪŵĳ,(4)ǳڱַҪһϵеĶ̾ͷ.ɼ,ͷƵȶӰƬṩһ൱õĸڽмֶ.
4  
ڱ,ӰƬȾ༭ĹƵźͨͷƵȡͷлƵһص,˶ԴƵźŸݾͷƵȽмķ.ʵƵźоͷбͽص,ǷֱƲʵӦľͷָ㷨ڴ˻˷ӳͷƵȵĶͷƽ֡³㷽.ʵӰƬеͼʾ,ͷƵȿԺܺõӰƬڽмҪ.
*оõ863߿ƼĿ͹ҡ211ѧƽĿ.
ͨѶϵˣѩ 100084,廪ѧѧ뼼ϵϢ
߼飺ѩ,1972,ʿ,ҪоΪݼ,Ӿ.
v,1940,,ʿʦ,ҪоΪӾ,ý弼ݼ.
ʷԪ,Ů,1967,ְʿ,ڣҪоΪֵ֧Эͬ,ý弼.
ߵλѩvʷԪ廪ѧѧ뼼ϵ100084
ѩvʷԪ(廪ѧܼϵͳصʵҡ100084)
οף
1Zhang H J, Kankanhalli A, Smoliar S W. Automatic partitioning of animate video. Technical Report, Institute of Systems Science, National University of Singapore, 1992
2Yeung M M, Yeo B L, Liu B. Extracting story units from long programs for video browsing and navigation. In IEEE Computer Society ed. Proceedings of the International Conference on Multimedia Computing and Systems. Los Alamitos IEEE Computer Society Press, 1996. 291305
3Irani M, Anandan P, Hsu S. Mosaic based representations of video sequences and their applications. In IEEE Computer Society ed. Proceedings of the 5th International Conference on Computer Vision. Los Alamitos IEEE Computer Society Press, 1995. 605611
4Ardizzone E, Cascia M L. Automatic video database indexing and retrieval. Multimedia Tools and Applications, 1997,4(2)2956
ոڣ1998-06-16޸ڣ1998-09-17
