|Table of Contents|

Advertisement Unit Segmentation Based on Fusion of Audio and Text


Research Field:
Publishing date:


Advertisement Unit Segmentation Based on Fusion of Audio and Text
ZHANG Yu-zhen123XIA Zhao-lin4WANG Jian-yu4DAI Yue-wei5
1.School of Electronic Engineering and Optoelectronic Technology,NUST,Nanjing 210094,China; 2.Jiangsu Key Laboratory of Spectral Imaging & Intelligent Sense,Nanjing 210094,China; 3.Key Laboratory of Photoelectronic Imaging Technology and System,Beijing Institute of Technology, Ministry of Education of China,Beijing 100081,China;4.School of Automation,NUST,Nanjing 210094,China; 5.Jiangsu University of Science and Technology,Zhenjiang 212003,China
advertisement units scene segmentation audio texts Gaussian mixture model entropy of segmentation audio change detection
Aiming at the problem of being difficult to segment ad units caused by the diversity of ad editorial methods,an ad unit segmentation algorithm based on fusing of audio data and text data is proposed.The audio data from ad video is modeled based on the Gaussian mixture model,and the audio change detection is realized based on the segmentation entropy.Along the time line,the ad unit boundary is detected for the first round by combining the audio change points and the text detection based on wavelet and support vector machine.The ad unit boundary is detected for the second round based on the time distance.Experiments prove that both the recall and precision of audio change detection are larger than 80% and both the recall and precision of ad unit segmentation are about 70%.


[1] Yeung M,Yeo B L,Liu B. Segmentation of video by clustering and graph analysis[J]. Computer Vision and Image Understanding, 1998, 71( 1) : 94-109.
[2] 张玉珍,王建宇,戴跃伟,等. 一种基于均值漂移的视频场景检测方法[J]. 中国图象图形学报,2010, 15( 2) : 314-320.
Zhang Yuzhen,Wang Jianyu,Dai Yuewei, et al. A video scene detection method based on mean shift[J]. Journal of Image and Graphics, 2010, 15( 2) : 314-320.
[3] 赵亚琴,周献中,何新. 一种层次的电影视频摘要生成方法[J]. 中国图象图形学报, 2007, 12( 8) : 1412-1417.
Zhao Yaqin,Zhou Xianzhong,He Xin. Automatically generating hierarchical summary for film video[J]. Journal of Image and Graphics, 2007, 12( 8) : 1412-1417.
[4] 程文刚,须德,郎从妍. 一种有效的视频场景检测方法[J]. 中国图象图形学报, 2004,9 ( 8) : 984-990.
Cheng Wengang,Xu De,Lang Congyan. An efficient method for video scene detection[J]. Journal of Image and Graphics, 2004,9 ( 8) : 984-990.
[5] 王学军,丁红涛,陈贺新. 一种基于镜头聚类的视频场景分割方法[J]. 中国图象图形学报,2007,12( 12) : 2127-2131.
Wang Xuejun,Ding Hongtao,Chen Hexin. A shot clustering based approach for scene segmentatin[J]. Journal of Image and Graphics, 2007, 12( 12) : 2127-2131.
[6] 李士进,郭跃飞,李昊,等. 新闻视频中广告片段精确定位方法研究[J]. 中国图象图形学报,2009,14 ( 7) : 1432-1439.
Li Shijin,Guo Yuefei,Li Hao,et al. Study on precise commercial segments location in news vide[J]. Journal of Image and Graphics, 2009, 14( 7) : 1432-1439.
[7] 张亮,朱振峰,赵耀,等. 基于镜头的鲁棒视频广告检测[J]. 智能系统学报, 2007,2 ( 2) : 83-88.
Zhang Liang,Zhu Zhenfeng,Zhao Yao, et al. Video commercial detection based on the robustness of shot[J]. Transactions on Intelligent System, 2007,2 ( 2) : 83-88.
[8] Huang Y P,Hsu L-W, Sandnes F E, et al. An intelligent subtitle detection model for locating television commercials [J]. IEEE Transactions on Systems,Man, and Cybernetics, 2007: 485-492.
[9] Rabiner L R. A tutorial on hidden Markov models and selected application in speech recognition[J]. Proc IEEE, 1989, 77( 2) : 257-286.
[10] C Shih-Sian,W Hsin-Min,F Hsin-Chia. BIC-based audio segmentation by divide-and-conquer [A]. Acoustics, Speech and Signal Processing, 2008. ICASSP 2008. IEEE International Conference on[C]. Las Vegas,USA: CRC Press, 2008: 4841-4844.
[11] 贾磊,穆向禺,徐波. 广播语音的音频边界分割[J].中文信息学报, 2002, 16( 1) : 37-42.
Jia Lei,Mu Xiangyu,Xu Bo. Broadcasting Sementation [J]. Journal of Chinese Information Processing, 2002, 16( 1) : 37-42.
[12] 王志明. 一种有效的音频分割算法[J]. 湖南理工学院学报( 自然科学版) , 2009, 22( 3) : 37-40.
Wang Zhiming. Audio segmentation based on layer entropy detection [J]. Journal of Hunan Institute of Science and Technology( Natural Sciences) ,2009,22 ( 3) : 37-40.
[13] 彭培华,曲波,陈荣胜. 基于支持向量机的小波域视频字幕检测与提取[J]. 华南理工大学学报( 自然科学版) , 2004, 32: 63-66.
Peng Peihua,Qu Bo,Chen Rongsheng. Video caption deteetion and extraetion in wavelet domain based on the support vector maehine[J]. Joumal of South China University of Technology ( Natural Science Edition) , 2004, 32: 63-66.


Last Update: 2012-10-12