Comparing images using joint histograms
Multimedia Systems - Special issue on video content based retrieval
A user attention model for video summarization
Proceedings of the tenth ACM international conference on Multimedia
Object Recognition as Machine Translation: Learning a Lexicon for a Fixed Image Vocabulary
ECCV '02 Proceedings of the 7th European Conference on Computer Vision-Part IV
CVPR '97 Proceedings of the 1997 Conference on Computer Vision and Pattern Recognition (CVPR '97)
Automatic image annotation and retrieval using cross-media relevance models
Proceedings of the 26th annual international ACM SIGIR conference on Research and development in informaion retrieval
Automatic Linguistic Indexing of Pictures by a Statistical Modeling Approach
IEEE Transactions on Pattern Analysis and Machine Intelligence
Tracking users' capture intention: a novel complementary view for home video content analysis
Proceedings of the 13th annual ACM international conference on Multimedia
Summarization of videotaped presentations: automatic analysis of motion and gesture
IEEE Transactions on Circuits and Systems for Video Technology
Semantic-event based analysis and segmentation of wedding ceremony videos
Proceedings of the international workshop on Workshop on multimedia information retrieval
IM(S)2: Interactive movie summarization system
Journal of Visual Communication and Image Representation
| Hi-index | 0.00 |
In this paper, we propose a video summarization system which takes into account users' individual preferences by using their personal photo libraries. Nowadays it is common, especially among people of younger generations, to store thousands of photos inside their PCs and manage them using software such as iPhoto and Picasa. These personal photo libraries contain rich information about the user's tastes, personalities, and lifestyles. Since still photos are in many aspects similar to video as a medium, we assume that these personal photo libraries can be used to estimate users' preferences on video summarization.Our system first divides a movie into short segments, and uses image classification techniques to judge whether each segment is meaningful to the user or not. If many photos with contents similar to the segment can be found in the user's photo library, the segment is judged as being "important" to the user. Conventional image classification techniques use public or commercial photo databases as training data, while our system uses personal photo libraries. This difference leads to the need of several modifications in the classification process.We have implemented a prototype version of our system, and have validated the effectiveness of our approach through evaluating both the accuracy of our image classification algorithm, and users' subjective satisfaction levels of the summarization results.