{"title":"基于关系图的视频抽象","authors":"Sulan Zhai, B. Luo, Jin Tang, Chunyan Zhang","doi":"10.1109/ICIG.2007.185","DOIUrl":null,"url":null,"abstract":"Video abstraction plays an important role in video browsing, video indexing, video retrieval and other video applications. In the paper, an automatic video abstraction method is developed based on relational graph representations. Firstly, a relational graph is constructed for all the frames in a video sequence. Secondly, the graph is partitioned into different connected subgraphs. Thirdly, Isomap is performed to reduce the dimensionality of the data set, and the output of Isomap is used as feature vector of the video frames. Lastly, a mixture model with model selection is introduced to generate the fine grain video abstraction with the centre of each clustering as the keyframes. Experiments are conducted on real world video sets with satisfying video abstraction results.","PeriodicalId":367106,"journal":{"name":"Fourth International Conference on Image and Graphics (ICIG 2007)","volume":"8 1","pages":"0"},"PeriodicalIF":0.0000,"publicationDate":"2007-08-22","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"6","resultStr":"{\"title\":\"Video Abstraction Based on Relational Graphs\",\"authors\":\"Sulan Zhai, B. Luo, Jin Tang, Chunyan Zhang\",\"doi\":\"10.1109/ICIG.2007.185\",\"DOIUrl\":null,\"url\":null,\"abstract\":\"Video abstraction plays an important role in video browsing, video indexing, video retrieval and other video applications. In the paper, an automatic video abstraction method is developed based on relational graph representations. Firstly, a relational graph is constructed for all the frames in a video sequence. Secondly, the graph is partitioned into different connected subgraphs. Thirdly, Isomap is performed to reduce the dimensionality of the data set, and the output of Isomap is used as feature vector of the video frames. Lastly, a mixture model with model selection is introduced to generate the fine grain video abstraction with the centre of each clustering as the keyframes. Experiments are conducted on real world video sets with satisfying video abstraction results.\",\"PeriodicalId\":367106,\"journal\":{\"name\":\"Fourth International Conference on Image and Graphics (ICIG 2007)\",\"volume\":\"8 1\",\"pages\":\"0\"},\"PeriodicalIF\":0.0000,\"publicationDate\":\"2007-08-22\",\"publicationTypes\":\"Journal Article\",\"fieldsOfStudy\":null,\"isOpenAccess\":false,\"openAccessPdf\":\"\",\"citationCount\":\"6\",\"resultStr\":null,\"platform\":\"Semanticscholar\",\"paperid\":null,\"PeriodicalName\":\"Fourth International Conference on Image and Graphics (ICIG 2007)\",\"FirstCategoryId\":\"1085\",\"ListUrlMain\":\"https://doi.org/10.1109/ICIG.2007.185\",\"RegionNum\":0,\"RegionCategory\":null,\"ArticlePicture\":[],\"TitleCN\":null,\"AbstractTextCN\":null,\"PMCID\":null,\"EPubDate\":\"\",\"PubModel\":\"\",\"JCR\":\"\",\"JCRName\":\"\",\"Score\":null,\"Total\":0}","platform":"Semanticscholar","paperid":null,"PeriodicalName":"Fourth International Conference on Image and Graphics (ICIG 2007)","FirstCategoryId":"1085","ListUrlMain":"https://doi.org/10.1109/ICIG.2007.185","RegionNum":0,"RegionCategory":null,"ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"","JCRName":"","Score":null,"Total":0}
Video abstraction plays an important role in video browsing, video indexing, video retrieval and other video applications. In the paper, an automatic video abstraction method is developed based on relational graph representations. Firstly, a relational graph is constructed for all the frames in a video sequence. Secondly, the graph is partitioned into different connected subgraphs. Thirdly, Isomap is performed to reduce the dimensionality of the data set, and the output of Isomap is used as feature vector of the video frames. Lastly, a mixture model with model selection is introduced to generate the fine grain video abstraction with the centre of each clustering as the keyframes. Experiments are conducted on real world video sets with satisfying video abstraction results.
求助内容：
应助结果提醒方式：
京公网安备 11010802042870号
