A bottom-up summarization algorithm for videos in the wild
Abstract Video summarization aims to provide a compact video representation while preserving the essential activities of the original video. Most existing video summarization approaches relay on identifying important frames and optimizing target energy by a global optimum solution. But global optimu...
Main Authors: | , , , , , |
---|---|
Format: | Article |
Language: | English |
Published: |
SpringerOpen
2019-02-01
|
Series: | EURASIP Journal on Advances in Signal Processing |
Subjects: | |
Online Access: | http://link.springer.com/article/10.1186/s13634-019-0611-y |
id |
doaj-0c1638635441494795a84506fc815e8f |
---|---|
record_format |
Article |
spelling |
doaj-0c1638635441494795a84506fc815e8f2020-11-25T03:35:36ZengSpringerOpenEURASIP Journal on Advances in Signal Processing1687-61802019-02-012019111110.1186/s13634-019-0611-yA bottom-up summarization algorithm for videos in the wildGang Pan0Yaoxian Zheng1Rufei Zhang2Zhenjun Han3Di Sun4Xingming Qu5College of Intelligence and Computing, Tianjin UniversityCollege of Intelligence and Computing, Tianjin UniversityBeijing Institute of Control and Electronics TechnologySchool of Electronic, Electrical and Communication Engineering, University of Chinese Academy of ScienceCollege of Intelligence and Computing, Tianjin UniversityCollege of Intelligence and Computing, Tianjin UniversityAbstract Video summarization aims to provide a compact video representation while preserving the essential activities of the original video. Most existing video summarization approaches relay on identifying important frames and optimizing target energy by a global optimum solution. But global optimum may fail to express continuous action or realistically validate how human beings perceive a story. In this paper, we present a bottom-up approach named clip growing for video summarization, which allows users to customize the quality of the video summaries. The proposed approach firstly uses clustering to oversegment video frames into video clips based on their similarity and proximity. Simultaneously, the importance of frames and clips is evaluated from their corresponding dissimilarity and representativeness. Then, video clips and frames are gradually selected according to their energy rank, until reaching the target length. Experimental results on SumMe dataset show that our algorithm can produce promising results compared to existing algorithms. Several video summarizations results are presented in supplementary material.http://link.springer.com/article/10.1186/s13634-019-0611-yVideo summarizationClip growingBottom-up |
collection |
DOAJ |
language |
English |
format |
Article |
sources |
DOAJ |
author |
Gang Pan Yaoxian Zheng Rufei Zhang Zhenjun Han Di Sun Xingming Qu |
spellingShingle |
Gang Pan Yaoxian Zheng Rufei Zhang Zhenjun Han Di Sun Xingming Qu A bottom-up summarization algorithm for videos in the wild EURASIP Journal on Advances in Signal Processing Video summarization Clip growing Bottom-up |
author_facet |
Gang Pan Yaoxian Zheng Rufei Zhang Zhenjun Han Di Sun Xingming Qu |
author_sort |
Gang Pan |
title |
A bottom-up summarization algorithm for videos in the wild |
title_short |
A bottom-up summarization algorithm for videos in the wild |
title_full |
A bottom-up summarization algorithm for videos in the wild |
title_fullStr |
A bottom-up summarization algorithm for videos in the wild |
title_full_unstemmed |
A bottom-up summarization algorithm for videos in the wild |
title_sort |
bottom-up summarization algorithm for videos in the wild |
publisher |
SpringerOpen |
series |
EURASIP Journal on Advances in Signal Processing |
issn |
1687-6180 |
publishDate |
2019-02-01 |
description |
Abstract Video summarization aims to provide a compact video representation while preserving the essential activities of the original video. Most existing video summarization approaches relay on identifying important frames and optimizing target energy by a global optimum solution. But global optimum may fail to express continuous action or realistically validate how human beings perceive a story. In this paper, we present a bottom-up approach named clip growing for video summarization, which allows users to customize the quality of the video summaries. The proposed approach firstly uses clustering to oversegment video frames into video clips based on their similarity and proximity. Simultaneously, the importance of frames and clips is evaluated from their corresponding dissimilarity and representativeness. Then, video clips and frames are gradually selected according to their energy rank, until reaching the target length. Experimental results on SumMe dataset show that our algorithm can produce promising results compared to existing algorithms. Several video summarizations results are presented in supplementary material. |
topic |
Video summarization Clip growing Bottom-up |
url |
http://link.springer.com/article/10.1186/s13634-019-0611-y |
work_keys_str_mv |
AT gangpan abottomupsummarizationalgorithmforvideosinthewild AT yaoxianzheng abottomupsummarizationalgorithmforvideosinthewild AT rufeizhang abottomupsummarizationalgorithmforvideosinthewild AT zhenjunhan abottomupsummarizationalgorithmforvideosinthewild AT disun abottomupsummarizationalgorithmforvideosinthewild AT xingmingqu abottomupsummarizationalgorithmforvideosinthewild AT gangpan bottomupsummarizationalgorithmforvideosinthewild AT yaoxianzheng bottomupsummarizationalgorithmforvideosinthewild AT rufeizhang bottomupsummarizationalgorithmforvideosinthewild AT zhenjunhan bottomupsummarizationalgorithmforvideosinthewild AT disun bottomupsummarizationalgorithmforvideosinthewild AT xingmingqu bottomupsummarizationalgorithmforvideosinthewild |
_version_ |
1724553483115298816 |