main.dvi

Size: px

Start display at page:

Download "main.dvi"

みいかしどり
4 years ago
Views:

1 DEIM Forum 2016 C Wikipedia Wikipedia () ( ),,, A Method of Assisting Movie Summarization by Aligning Plot Sentences and Shots ueshan LI, Kentaro DOMOTO, and Takehito UTSURO Grad. Sch. of Systems and Information Engineering, University of Tsukuba, Tsukuba Japan Faculty of Engineering, Information and Systems, University of Tsukuba, Tsukuba Japan 1. 1 Wikipedia Wikipedia () ( ) 2. 1 Wikipedia ( (plot)) [1] 2

2 1 2 Wikipedia () ( ) 3. Wikipedia Wikipedia 1 1 Wikipedia

The seda ve eventually makes her fall asleep on a

reporter working for an American news service based

Not recognizing her, he offers her money so she can

she later calls herself) refuses to cooperate.

The next morning, Joe, having already slept through

hurries off to work, leaving her s ll asleep.

Hennessy (Hartley Power), asks why Joe is late, Joe

3 The seda ve eventually makes her fall asleep on a bench, where Joe Bradley, an expatriate American reporter working for an American news service based in Rome, finds her. Not recognizing her, he offers her money so she can take a taxi home, but a very woozy "Anya Smith" (as she later calls herself) refuses to cooperate. Joe finally decides, for safety's sake, to let her spend the night in his apartment. He is amused by her regal manner, but less so when she appropriates his bed. He transfers her to a couch. The next morning, Joe, having already slept through the interview Princess Ann was scheduled to give, hurries off to work, leaving her s ll asleep. When his editor, Mr. Hennessy (Hartley Power), asks why Joe is late, Joe lies, claiming to have a ended the press conference for the princess. Joe makes up details of the alleged interview un l Hennessy informs him that the event had been canceled because the princess had suddenly fallen ill. Joe sees a picture of her and realizes who is in his apartment. 4 3 () Wikipedia

4 [1] [6] [6] Wikipedia () [6] s i t j f fus (s i,t j) f fus (s i,t j)=f id (s i,t j)+α f subtt (s i,t j) f id (s i,t j) f subtt (s i,t j) s i t j (6. 1 ) () (6. 2 ) α [6] [6] 2http://mklab.iti.gr/project/video-shot-segm 3 [1] 88.7% ( [6] 83 ) 6. 1 s i t j f id (s i,t j) c( C C ) d( D D ) ( 1 ( c d ) align(c, d) = 0 () () ( ) align(c, d) s i t j f id (s i,t j) f id (s i,t j)= j+r k=j r c C j d D i align(c, d) I(c) t j r C j s i D i align(c, d) c( C j) d( D i) f id (s i,t j) c c n FT(c) IDF(inverse document frequency) I(c )= log max c C nft(c) log n FT(c ) () f subtt (s i,t j) () t j subtt s i v t j subtt w f subtt (s i,t j) word-match(v, w) v w 1 f subtt (s i,t j)= word-match(v, w) v s i w subtt t j ( 1 (v = w) word-match(v, w) = 0 (v = w)

5 % % % 4 Wikipedia ( ) 6. 3 [6] () ( ) ( 20 ) 73% [6]

6 Wikipedia ( 1.9%) ( 2.7%) ( 2.8%) 4 8. [2] [9] [3] [8] [7] [6] Wikipedia [4] [1] E. Apostolidis and V. Mezaris. Fast shot segmentation combining global and local visual descriptors1. In Proc. ICASSP, pp , [2],.., Vol DBS 132, pp , [3] C. Liang, Y. Zhang, J. Cheng, C. u, and H. Lu. A novel role-based movie scene segmentation method. In Advances in Multimedia Information Processing PCM2009, Vol of LNCS, pp Springer, [4],.. 3, pp , [5] P. Sidiropoulos, V. Mezaris, I. Kompatsiaris, H. Meinedo, M. Bugalho, and I. Trancoso. Temporal video segmentation to scenes using high-level audiovisual features. IEEE Transactions on Circuits and Systems for Video Technology, Vol. 21, No. 8, pp , [6] M. Tapaswi, M. Bäuml, and R. Stiefelhagen. Aligning plot synopses to videos for story-based retrieval. International Journal of Multimedia Information Retrieval, Vol. 4, No. 1, pp. 3 16, [7] T. Tsoneva, M. Barbieri, and H. Weda. Automated summarization of narrative video on a semantic level. In Proc. Semantic Computing, pp , [8] H. Yi, D. Rajan, and L.-T. Chia. Semantic video indexing and summarization using subtitles. In Advances in Multimedia Information Processing PCM2004, Vol of LNCS, pp Springer, [9],,.., Vol HCI 124, pp , Wikipedia [6] [5]

高2SL高1HL 文法後期後半_テキスト-0108.indd

高2SL高1HL 文法後期後半_テキスト-0108.indd 第 20 講関係詞 3 ポイント 1 -ever 2 3 ポイント 1 複合関係詞 (-ever) ever whoever whatever whichever whenever wherever You may take whoever wants to go. Whenever she comes, she brings us presents. = no matter whoever =