Video Pulses: User-based modeling of interesting video segments. Avlonitis, M. & Chorianopoulos, K. Advances in Multimedia, 2014. Paper Paper doi abstract bibtex We present a user-based method that detects regions of interest within a video, in order to provide video skims and video summaries. Previous research in video retrieval has focused on content-based techniques, such as pattern recognition algorithms that attempt to understand the low-level features of a video. We are proposing a pulse modeling method, which makes sense of a web video by analyzing users Replay interactions with the video player. In particular, we have modeled the user information seeking behavior as a time series and the semantic regions as a discrete pulse of fixed width. Then, we have calculated the correlation coefficient between the dynamically detected pulses at the local maximums of the user activity signal and the pulse of reference. We have found that users Replay activity significantly matches the important segments in information-rich and visually complex videos, such as lecture, how-to, and documentary. The proposed signal processing of user activity is complementary to previous work in content-based video retrieval and provides an additional user-based dimension for modeling the semantics of a social video on the Web.
@Article{Avlonitis_2014,
author = {Avlonitis, Markos and Chorianopoulos, Konstantinos},
title = {{Video Pulses: User-based modeling of interesting video segments}},
journal = {Advances in Multimedia},
year = {2014},
pages = {1--9},
abstract = {We present a user-based method that detects regions of interest within a video, in order to provide video
skims and video summaries. Previous research in video retrieval has focused on content-based techniques, such as pattern
recognition algorithms that attempt to understand the low-level features of a video. We are proposing a pulse modeling method,
which makes sense of a web video by analyzing users Replay interactions with the video player. In particular, we have modeled
the user information seeking behavior as a time series and the semantic regions as a discrete pulse of fixed width. Then, we
have calculated the correlation coefficient between the dynamically detected pulses at the local maximums of the user activity
signal and the pulse of reference. We have found that users Replay activity significantly matches the important segments in
information-rich and visually complex videos, such as lecture, how-to, and documentary. The proposed signal processing of user
activity is complementary to previous work in content-based video retrieval and provides an additional user-based dimension for
modeling the semantics of a social video on the Web.},
doi = {10.1155/2014/712589},
url_Paper={Avlonitis_2014.pdf},
keywords = {analytics,human-computer interaction,implicit,information retrieval,interaction,multimedia,region of
interest,semantics,signal processing,time-series,user modeling,video,video lecture},
mendeley-tags = {analytics,human-computer interaction,information retrieval,multimedia,semantics,signal
processing,time-series,video lecture},
url = {http://www.hindawi.com/journals/am/2014/712589/},
}
Downloads: 0
{"_id":"ZHDe3saht4KBdwXxa","authorIDs":["545a0c50b43425b772000bdf","5deb8c7eb62591df0100006e","5dfde29ddd49fdde01000068","5e10db5645c12cde01000005","5e17e2935eee58df01000008","5e2ad1f6638921df010000a7","5j7JZm4HkoJLGbKLB","7h7QT2JDiSYesYpft","LNhHkxcu7yD7GLnjT","PPBe4vuty3QxtdXvw","h3d4b9cPXz2czRy28","i4z4B7ziYYMret35J","j5HKKSGowdyysMjZj"],"author_short":["Avlonitis, M.","Chorianopoulos, K."],"bibbaseid":"avlonitis-chorianopoulos-videopulsesuserbasedmodelingofinterestingvideosegments-2014","bibdata":{"bibtype":"article","type":"article","author":[{"propositions":[],"lastnames":["Avlonitis"],"firstnames":["Markos"],"suffixes":[]},{"propositions":[],"lastnames":["Chorianopoulos"],"firstnames":["Konstantinos"],"suffixes":[]}],"title":"Video Pulses: User-based modeling of interesting video segments","journal":"Advances in Multimedia","year":"2014","pages":"1–9","abstract":"We present a user-based method that detects regions of interest within a video, in order to provide video skims and video summaries. Previous research in video retrieval has focused on content-based techniques, such as pattern recognition algorithms that attempt to understand the low-level features of a video. We are proposing a pulse modeling method, which makes sense of a web video by analyzing users Replay interactions with the video player. In particular, we have modeled the user information seeking behavior as a time series and the semantic regions as a discrete pulse of fixed width. Then, we have calculated the correlation coefficient between the dynamically detected pulses at the local maximums of the user activity signal and the pulse of reference. We have found that users Replay activity significantly matches the important segments in information-rich and visually complex videos, such as lecture, how-to, and documentary. The proposed signal processing of user activity is complementary to previous work in content-based video retrieval and provides an additional user-based dimension for modeling the semantics of a social video on the Web.","doi":"10.1155/2014/712589","url_paper":"Avlonitis_2014.pdf","keywords":"analytics,human-computer interaction,implicit,information retrieval,interaction,multimedia,region of interest,semantics,signal processing,time-series,user modeling,video,video lecture","mendeley-tags":"analytics,human-computer interaction,information retrieval,multimedia,semantics,signal processing,time-series,video lecture","url":"http://www.hindawi.com/journals/am/2014/712589/","bibtex":"@Article{Avlonitis_2014,\n author = {Avlonitis, Markos and Chorianopoulos, Konstantinos},\n title = {{Video Pulses: User-based modeling of interesting video segments}},\n journal = {Advances in Multimedia},\n year = {2014},\n pages = {1--9},\n abstract = {We present a user-based method that detects regions of interest within a video, in order to provide video\nskims and video summaries. Previous research in video retrieval has focused on content-based techniques, such as pattern\nrecognition algorithms that attempt to understand the low-level features of a video. We are proposing a pulse modeling method,\nwhich makes sense of a web video by analyzing users Replay interactions with the video player. In particular, we have modeled\nthe user information seeking behavior as a time series and the semantic regions as a discrete pulse of fixed width. Then, we\nhave calculated the correlation coefficient between the dynamically detected pulses at the local maximums of the user activity\nsignal and the pulse of reference. We have found that users Replay activity significantly matches the important segments in\ninformation-rich and visually complex videos, such as lecture, how-to, and documentary. The proposed signal processing of user\nactivity is complementary to previous work in content-based video retrieval and provides an additional user-based dimension for\nmodeling the semantics of a social video on the Web.},\n doi = {10.1155/2014/712589},\n url_Paper={Avlonitis_2014.pdf},\n keywords = {analytics,human-computer interaction,implicit,information retrieval,interaction,multimedia,region of\ninterest,semantics,signal processing,time-series,user modeling,video,video lecture},\n mendeley-tags = {analytics,human-computer interaction,information retrieval,multimedia,semantics,signal\nprocessing,time-series,video lecture},\n url = {http://www.hindawi.com/journals/am/2014/712589/},\n}\n\n","author_short":["Avlonitis, M.","Chorianopoulos, K."],"key":"Avlonitis_2014","id":"Avlonitis_2014","bibbaseid":"avlonitis-chorianopoulos-videopulsesuserbasedmodelingofinterestingvideosegments-2014","role":"author","urls":{" paper":"https://pdf.epidro.me/Avlonitis_2014.pdf","Paper":"http://www.hindawi.com/journals/am/2014/712589/"},"keyword":["analytics","human-computer interaction","implicit","information retrieval","interaction","multimedia","region of interest","semantics","signal processing","time-series","user modeling","video","video lecture"],"metadata":{"authorlinks":{"chorianopoulos, k":"https://pdf.epidro.me/"}},"downloads":0,"html":""},"bibtype":"article","biburl":"https://pdf.epidro.me/publications.bib","creationDate":"2014-12-29T22:19:46.734Z","downloads":0,"keywords":["analytics","human-computer interaction","implicit","information retrieval","interaction","multimedia","region of interest","semantics","signal processing","time-series","user modeling","video","video lecture"],"search_terms":["video","pulses","user","based","modeling","interesting","video","segments","avlonitis","chorianopoulos"],"title":"Video Pulses: User-based modeling of interesting video segments","year":2014,"dataSources":["7y6dBnX9BfJmF3as3"]}