媒體豐富因素對視訊摘錄代表性之影響

The Effect of Media Richness Factors on Representativeness for Video Skim

盧增錄(Lu-Tseng Lu)

指導教授 ： 孫惠民

長榮大學/資訊暨工程學院/資訊管理學系(所)/碩士(2010年)

https://doi.org/10.6833/CJCU.2010.00077

摘要

本研究以媒體豐富理論中的媒介個別化焦點與多重線索因素，探討視訊摘錄代表性之影響，並分別驗證其對視訊摘錄代表性之影響。其中，媒介個別化焦點代表在不同類型的電影有著不同的焦點，而多重線索則代表同時具有視訊、音訊及文字。在代表性方面，我們分為兩種指標，分別為不確定性與理解程度，透過主觀的視訊品質量測方法，來評估摘錄的代表性。本研究共分為兩個部份，第一部分為媒介個別化焦點因素在不同類型電影的影響，結果顯示在恐怖片與動作片其視訊摘錄代表性皆有顯著差異，而集中結局的方式可以顯著的提升其視訊摘錄代表性。在第二部份則是以多重線索因素作為視訊摘錄策略，並與第一部份的媒介個別化焦點的集中結局比較，結果顯示使用多重線索因素較單獨使用媒介個別化焦點因素，更具視訊摘錄之代表性。因此，在本研究的三種不同類型的影片，以影片內容最常出現之字幕所呈現的視訊摘錄，最能提昇視訊摘錄之代表性。

關鍵字

內容特徵 ； 視訊摘錄 ； 媒體豐富理論

並列摘要

In this study, the representativeness of skimmed video is based on the personality focus of the medium and the multiple cues from the media richness theory, and we explore the effect of the representativeness on skimmed video. The personality focus of the medium means that there is different focus among different types of movies, while the multiple cues consist of video, audio, and subtitles. For the representativeness, we use two indicators, certainty and comprehension, to evaluate the skimmed video measured by a subjective quality assessment method. We design two experiments. For experiment 1, the personality focus of the medium has significantly different from movie types for the representativeness. The experimental results show that skimming most ending part can significantly improve the representativeness for horror film and action film. For experiment 2, skimmed video with multiple cues can more significantly improve the representativeness than that only with used the personality focus of the medium. Therefore, for the three types of movies, the skimmed video used the frequent words’ stem from subtitles has batter the representativeness.

並列關鍵字

Content Characteristic ； Video Skim ； Media Richness Theory

參考文獻

52. 黃志偉 (2009), 「探討劇情片電影視訊摘錄之影響」. 長榮大學 資訊管理學系 碩士論文.

2. Daft, R.L. and R.H. Lengel (1986), Organizational Information Requirements, Media Richness and Structural Design. Management Science, 1986. Vol.32(No.5): p. 554-571.

3. Shi, L., M.R. Lyu, and I. King (2004), Video summarization by spatial-temporal graph optimization, in Circuits and Systems, 2004. ISCAS '04. Proceedings of the 2004 International Symposium on. 2004. p. 197-200

4. Gong, Y. and X. Liu (2003), Video summarization and retrieval using singular value decomposition. ACM Multimedia Syst, 2003. Vol.9(No.2): p. 157-168.

6. Gong, Y.H. (2003), Summarizing audio-visual contents of a video program. EURASIP J. Signal Process Special Issue on Unstructured Info. Manage Multimedia Data Sources, 2003: p. 160-169.