首页> 外文会议>Active media technology >Fractal Based Video Shot Cut/Fade Detection and Classification
【24h】

Fractal Based Video Shot Cut/Fade Detection and Classification

机译:基于分形的视频镜头切换/淡入淡出检测与分类

获取原文
获取原文并翻译 | 示例

摘要

Video segmentation plays an important role in video indexing, content-based video coding and retrieval. In this paper, we propose a new method for cut and fade detection using fractal dimension. We also classify frames into three categories: "CUT", "FADE IN/OUT", and "None SHOT". To test our method, we used 20 videos which contain more than 33,000 frames in different subjects, including different type of shot boundaries. It was also successfully compared to two other methods of shot boundary detection. Results from experiments depict the improved precision and recall of the proposed method in recognition of the fade in/out frames our database.
机译:视频分割在视频索引,基于内容的视频编码和检索中起着重要作用。在本文中,我们提出了一种使用分形维数进行剪切和褪色检测的新方法。我们还将帧分为三类:“剪切”,“淡入/淡出”和“无显示”。为了测试我们的方法,我们使用了20个视频,其中包含33,000多帧不同主题的画面,包括不同类型的镜头边界。它还成功地与镜头边界检测的其他两种方法进行了比较。实验结果表明,在识别数据库淡入/淡出帧时,该方法具有更高的精度和召回率。

著录项

相似文献

  • 外文文献
  • 中文文献
  • 专利
获取原文

客服邮箱:kefu@zhangqiaokeyan.com

京公网安备:11010802029741号 ICP备案号:京ICP备15016152号-6 六维联合信息科技 (北京) 有限公司©版权所有
  • 客服微信

  • 服务号