演員戲份實驗結果分析

（a）武俠動作片 （b）動畫特效片

（c）文藝愛情片 （d）戰爭格鬥片 圖三十五、右樣板判斷正確之特寫鏡頭

0.00%

10.00%

20.00%

30.00%

40.00%

50.00%

60.00%

70.00%

80.00%

90.00%

100.00%

2 3 4 5 6 7

叢集個數

Precision

沒有背景補償 全面背景補償 採樣背景補償

圖三十六、準確率比較圖

0.00%

5.00%

10.00%

15.00%

20.00%

25.00%

30.00%

35.00%

40.00%

45.00%

50.00%

2 3 4 5 6 7

叢集個數

Recall

沒有背景補償 全面背景補償 採樣背景補償

圖三十七、回復率比較圖

7. 結論及未來的工作

本文提出一種以樣板比對為基礎的特寫鏡頭偵測方法，能夠有效的偵測到特寫鏡 頭，並將其做叢集處理來進行主角的戲份比重計算，並進一步自動合成電影摘要。希望 能藉由電影摘要的自動合成，能夠讓使用者快速瞭解到一部電影的內涵，充分發揮電影 資料庫的典藏功能。

我們未來的首要工作為提昇鏡頭叢集技術的準確率，並進行大規模實驗。經由提昇 叢集技術我們希望除了可以識別演員戲份外，更能進一步判別男演員與女演員，如此便 能識別出男主角與女主角，使得電影摘要的合成能更適合電影真正內涵。若能進行大規 模實驗，就能藉由實驗的結果修正我們的參數設定，進一步增加我們的準確率與回復 率，相對的電影自動化摘要的效果和可靠性也會相對地提高。

在電影預告片的合成方面，為了能讓預告片更加的生動活潑，除了擷取特寫鏡頭的 場景之外，未來希望能加入電影音效的輔助，例如穿插包含重低音的場景或是擁有環繞 音效的場景。我們將對不同的電影類型（文藝、動作、科幻、恐怖、喜劇）個別處理，

找出各類型中適合的摘要類型。例如文藝片可需要主要角色的特寫鏡頭場景與含背景音 樂場景搭配；而動作或科幻片則是需要主要角色的特寫鏡頭場景與包含重低音的場景或 是擁有環繞音效的場景作搭配。如此一來電影自動化摘要系統就能更符合使用者的需 求，電影自動化摘要系統將更具實用價值性。

多媒體內容描述介面（MPEG-7）包含了描述工具與描述定義語言（DDL）。因此 我們希望未來能使用描述定義語言來制定 MPEG-4 電影摘要特徵綱要，並且遵循 MPEG-7 電影摘要特徵綱要所定義的結構來描述電影的內涵。

參考文獻

[1] ISO/IEC 14496-2:1998, “Information Technology-Generic Coding of Audio-Visual Objects.”

[2] ISO/IEC JTC1/SC29/WG11:2002, “Coding of Moving Pictures and Audio.”

[3] Berna Erol and Faouzi Kossentini, “Automatic Key Video Object Plane Selection Using the Shape Information in the MPEG-4 Compressed,” IEEE Trans. on Circuits and Systems for Video Technology, Vol.

2, No.2, pp. 129-138, 2000.

[4] Berna Erol and Faouzi Kossentini, “Video Object Summarization in The MPEG-4 Compressed Domain,”

in International Conference on Acoustics, Speech, and Signal Processing, pp. 2027-2030, 2000

[5] Tsuyoshi Moriyama and Masao Sakauchi, “Video Summarization Based on the Psychological Content in the Track Structure,” In Processing of ACM Multimedia Workshop, pp. 191-194, 2000.

[6] Microsoft, JTC1/SC29/WG11:2000, “MPEG-4 Video Encoder/Decoder..”

[7] Candemir Toklu, Shih-Ping Liou, and Madirakshi Das, “Video Abstract: A Hybrid Approach to Generate Semantically Meaningful Video Summaries,” In Processing of IEEE International Conference on Multimedia and Expo, Vol. 3, pp. 1333-1336, 2000.

[8] Yihong Gong and Xin Liu, “Generating Optimal Video Summaries,” In Processing of IEEE International Conference on Multimedia and Expo Vol. 3, pp. 1559-1562, 2000.

[9] Yihong Gong and Xin Liu, “Video Summarization with Minimal Visual Content Redundancies,” In Processing of IEEE International Conference on Multimedia and Expo, Vol. 3, pp.

362-365, 2001.

[10] Nuno Vasconcelos and Andrew Lippman, “Bayesian Modeling of Video Editing and Structure: Semantic Features for Video Summarization and Browsing,” In Processing of IEEE International Conference on Image Processing, Vol. 3, pp. 153-157, 1998.

[11] S. W. Smoliar and H. Zhang, “Content-based video indexing and retrieval,” IEEE Multimedia Magazine, pp. 62–72, 1994.

[12] Weiping Li, “Overview of Fine Granularity Scalability in MPEG-4 Video Standard,”

IEEE Transactions on Circuits and Systems for Video Technology, Vol. 11, No. 3, pp.

301-317, 2001.

[13] N. Brady, F. Bossen, N. Murphy, “Context-Based Arithmetic Encoding of 2D Shape Sequences,” in IEEE International Conference on Image Processing, pp. 29-32 1997.

[14] A. Katsaggelos, et al. “MPEG-4 and rate-distortion based shape coding techniques,”

Proceedings of the IEEE, pp. 1126-1154, 1998.

[15] J. Ostermann, “Coding of arbitrarily shaped video objects with binary and greyscale alpha maps: What can MPEG-4 do for you?” in Processing of IEEE International Symposium on Circuits and Systems. Vol. 5, pp. 273-276 1998.

[16] K. Changick and H. Jenq-Neng, “An Integrated Scheme for Object-Based Video Abstraction.” in Processing of ACM Multimedia Conf., pp. 303-311, 2000.

[17] Mei-Juan Chen, Yuan-Pin Hsieh, and Yu-Pin Wang, “Multi-Resolution Shape Coding Algorithm For MPEG-4,” IEEE Transactions on Consumer Electronics, Vol. 46, No. 3, 2000.

[18] David Saxe and Richard Foulds, “Toward Robust Skin Identification in Video Images,”

in Proceedings of the Second International Conference on Automatic Face and Gesture Recognition, pp. 379-384, 1996

[19] S. Ahmad, “A usable real-time 3d hand tracker,” in Processing of Conference Record of the Twenty-Eighth Asilomar Conference on Signals, Systems and Computers , pp.

1257-1261, 1994

[20] A. K. Jain, M. N. Murty, and P. J. Flynn, “Data clustering: a review,” ACM Computing Surveys , Vol. 31, NO. 3, 1999.

[21] JungHwan Oh and Hua K.A, ” An Efficient Technique for Summarizing Videos Using Visual Contents,” in Processing of IEEE International Conference on Multimedia and Expo, Vol. 2, pp.

1167-1170, 2000.

[22] Minerva M. Yeung and Boon-Lock Yeo, “Video visualization for compact presentation

and fast browsing of pictorial content,” IEEE Transactions on Circuits and Systems for Video Technology, Vol. 7, No. 5, October 1997

[23] D. DeMenthon,V. Kobla, and D. Doermann, “Video summarization by curve simplification,” In Proc. of ACM Int'l Conf. on Multimedia, pp. 211-218, Auguest 1998 [24] S.Uchihash, J. Foote, A. Girgensohn, and J. Boreczky. “Video manga: Generating

semantically meaningful video summaries,” In Proc. of ACM Int'l Conf. on Multimedia, pp. 383-392, Oct. 1999

[25] S.Uchihash and J. Foote, “Summarizing video using a shot importance measure and frame-packing algorithm,” In Proc. of ICASSP '99, Vol.6, pp.3041-3344, 1999

[26] Q. Hunag, Z. Lui, and A. Rosenberg, “Automated semantic structure reconstruction and representation generation for broadcast news,” In Proc. SPIE Conference on Storage and Retrieval for Image and Video database VII, Vol. 3656, pp. 50-62, 1999.

[27] M. Christel, et al. “Informedia digital video library,” Communication of the ACM Vol.

38 No. 4 pp. 57-58 1995.

[28] M. Smith and T. Kabade, “Video skimming and characterization through the combination of image and language understanding techniques,” In Proc. of Computer Vision and Pattern Recognition, pp. 775-781, 1997.

[29] M. Christel, et al. “Evolving video skims into useful multimedia abstractions,” In Proc.

of Human Factors in Computing System, CHI 98, pp. 171-178, 1998.

[30] R. Lienhart, “Abstracting home video automatically,” In Proc. ACM Multimedia 99

（Part2）, pp.37-40, 1999.

[31] Rainer Lienhart, Silvia Pfeiffer, and Wolfgang Effelsberg, “Video Abstracting,”

Communications of the ACM, Vol. 40, No. 12, pp. 55-62, 1997.

[32] Konigsberg, I, The Complete Film Dictionary,2 ed., Penguin Reference, 1997.

[33] Katz, E., The Film Encyclopedia, 4 ed., Harper Collins, 2001.