Abstract
Nowadays, content-based retrieval of video material is based on the availability of meta-data linked to it. Current approaches to automatically extract these data start from a temporal segmentation of the audio-visual material, that is, a location of the camera shot transitions. Although abrupt transition detection is a problem almost solved, success on gradual transition detection is still very low. We present a detector for all types of gradual transitions (chromatic, geometric, and mixed), based on in depth modelling of the patterns that these transitions generate on a specific frame distance. Results are presented over a representative sample of more than 250 gradual transitions, which belong to a significant part (200') of the MPEG-7 testing material, achieving a recall (percentage of effects correctly detected) and a precision (percentage of non-false positives in the set of detection) notably higher than the ones reported so far.
Sign-in/Register to access full text options 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a callDisclaimer: All third-party content on this website/platform is and will remain the property of their respective owners and is provided on "as is" basis without any warranties, express or implied. Use of third-party content does not indicate any affiliation, sponsorship with or endorsement by them. Any references to third-party content is to identify the corresponding services and shall be considered fair use under The CopyrightLaw.
