VideoMontage_CVPR06 - Space-Time Video Montage Hong-Wen...

Info iconThis preview shows pages 1–2. Sign up to view the full content.

View Full Document Right Arrow Icon

Info iconThis preview has intentionally blurred sections. Sign up to view the full version.

View Full DocumentRight Arrow Icon
This is the end of the preview. Sign up to access the rest of the document.

Unformatted text preview: Space-Time Video Montage Hong-Wen Kang * Yasuyuki Matsushita Xiaoou Tang Xue-Quan Chen University of Science and Technology of China Hefei, P.R.China { hwkang@mail.,chenxq@ } ustc.edu.cn Microsoft Research Asia Beijing , P.R.China { yasumat,xitang } @microsoft.com Abstract Conventional video summarization methods focus pre- dominantly on summarizing videos along the time axis, such as building a movie trailer. The resulting video trailer tends to retain much empty space in the background of the video frames while discarding much informative video content due to size limit. In this paper, we propose a novel space- time video summarization method which we call space-time video montage. The method simultaneously analyzes both the spatial and temporal information distribution in a video sequence, and extracts the visually informative space-time portions of the input videos. The informative video portions are represented in volumetric layers. The layers are then packed together in a small output video volume such that the total amount of visual information in the video volume is maximized. To achieve the packing process, we develop a new algorithm based upon the first-fit and Graph cut op- timization techniques. Since our method is able to cut off spatially and temporally less informative portions, it is able to generate much more compact yet highly informative out- put videos. The effectiveness of our method is validated by extensive experiments over a wide variety of videos. 1. Introduction The rapid increase of the amount of on-line and off-line video data necessitates development of efficient tools for fast video browsing. Video summarization [ 6 , 14 , 13 ] is one approach toward tackling this problem, in that it auto- matically creates a short version of the original input video. Summarized video content is important for many practical applications such as archiving 24-hour security videos and providing easy access to large sets of digital video docu- mentaries. This paper addresses the problem of automatically syn- * This work was done while the first author was visiting Microsoft Re- search Asia. Figure 1. Idea of the space-time video montage. thesizing a new short/small video from a long input video sequence by extracting and fusing the space-time informa- tive portions of the input video. Unlike prior video summa- rization methods, our method is not limited to the frame- basis, but uses an arbitrary space-time volume that is ex- tracted as the informative video portion. The extracted 3D informative video portions are packed together in the output video volume in a way in which the total visual informa- tion is maximized. This approach generates a compact yet highly informative video that tries to retain most informa- tive portions of the input videos....
View Full Document

This note was uploaded on 06/13/2011 for the course CAP 6412 taught by Professor Staff during the Spring '08 term at University of Central Florida.

Page1 / 8

VideoMontage_CVPR06 - Space-Time Video Montage Hong-Wen...

This preview shows document pages 1 - 2. Sign up to view the full document.

View Full Document Right Arrow Icon
Ask a homework question - tutors are online