Search In this Thesis
   Search In this Thesis  
العنوان
Enhanced Video Summarization and Key Frame Extraction Using Visual Features \
المؤلف
Mohamed, Karim Mohamed Mahmoud.
هيئة الاعداد
مشرف / كريم محمد محمود محمد
مشرف / محمد عبد الحميد إسماعيل
مشرف / نجية محمد غانم
مناقش / مجدى حسين ناجى
مناقش / ياسر السنباطى
الموضوع
Computer Science.
تاريخ النشر
2012.
عدد الصفحات
123 p. :
اللغة
الإنجليزية
الدرجة
ماجستير
التخصص
هندسة النظم والتحكم
تاريخ الإجازة
1/12/2012
مكان الإجازة
جامعة الاسكندريه - كلية الهندسة - هندسة الحاسب والنظم
الفهرس
Only 14 pages are availabe for public view

from 138

from 138

Abstract

The fast evolution of digital video has brought many new multimedia applications and, as a consequence, has increased the amount of research into new technologies that aim at improving the effectiveness and efficiency of video acquisition, archiving, cataloging and indexing, as well as increasing the usability of stored videos. Among possible research areas, video summarizalion is an important topic that potentially enables faster browsing of large video collections and also more efficient content indexing and access. Essentially, this research area consists of automatically generating a short summary of a video by selecting a set of salient images called key frames to represent the video content.
In this thesis, we propose two new enhanced approaches for the production of static video summaries. The first approach is a clip-based video summarization approach that combines both color features and texture features of the video frames using a modified density based clustering algorithm to generate the key frames. While the second approach is a shot-based video summarization approach using a two stage method based on video shots detection from color features in the first stage and the generation of key frames by extracting texture features of shots representatives using a nearest neighbor graph in the second stage.
As an additional contribution, an enhanced evaluation method is developed for assessing the quality of static video summaries. In this evaluation methodology, video summaries are manually created by users in previous recent study. Then, several user-created summaries are compared both to our approach and also to a number of different techniques in the literature using both color and texture features. Experimental evaluation conducted on 50 different videos shows that the two proposed video summarization approaches provide static video summaries with higher quality relative to the other approaches to which they were compared.