UNIGE document Scientific Article
previous document  unige:47463  next document
add to browser collection
Title

Information-theoretic temporal segmentation of videos and applications : multiscale keyframe selection and transition detection

Authors
Published in Multimedia Tools and Applications. 2006, vol. 30, p. 273-288
Abstract The first step in the analysis of video content is the partitioning of a long video sequence into short homogeneous temporal segments. The homogeneity property ensures that the segments are taken by a single camera and represent a continuous action in time and space. These segments can then be used as atomic temporal components for higher level analysis like browsing, classification, indexing and retrieval. The novelty of our approach is to use color information to partition the video into segments dynamically homogeneous using a criterion inspired by compact coding theory. We perform an information-based segmentation using a Minimum Message Length (MML) criterion and minimization by a Dynamic Programming Algorithm (DPA). We show that our method is efficient and robust to detect all types of transitions in a generic manner. A specific detector for each type of transition of interest therefore becomes unnecessary. We illustrate our technique by two applications: a multiscale keyframe selection and a generic shot boundaries detection.
Keywords Content-based video analysisTemporal segmentationKeyframe selectionDetection of shot boundaries
Identifiers
Full text
Structures
Research groups Computer Vision and Multimedia Laboratory
Multimodal Interaction Group
Viper group
Citation
(ISO format)
JANVIER, Bruno et al. Information-theoretic temporal segmentation of videos and applications : multiscale keyframe selection and transition detection. In: Multimedia Tools and Applications, 2006, vol. 30, p. 273-288. https://archive-ouverte.unige.ch/unige:47463

141 hits

49 downloads

Update

Deposited on : 2015-03-03

Export document
Format :
Citation style :