VIDEO HIGHLIGHT DETECTION BASED USER RETENTION CURVES
Zhu, Jiayu
Loading…
Permalink
https://hdl.handle.net/2142/124783
Description
Title
VIDEO HIGHLIGHT DETECTION BASED USER RETENTION CURVES
Author(s)
Zhu, Jiayu
Issue Date
2023-05-01
Keyword(s)
video highlight; retention curve; MLP
Abstract
Automatically detecting highlight moments in videos is a highly valuable common need in the time of the explosive growth of the number of videos. In this paper, we create a large-scale dataset, including the new retention curves recorded by the content provider websites to locate the highlight moments within a specific video. Besides the harvested dataset, we also propose a model, added retention curves as labels for training, to analyze the highlight moments. The video/audio features of the video chunks with the most/least audience are extracted and trained through a MLP model. Experiments show that our model has good potential to reach decent accuracy.
Use this login method if you
don't
have an
@illinois.edu
email address.
(Oops, I do have one)
IDEALS migrated to a new platform on June 23, 2022. If you created
your account prior to this date, you will have to reset your password
using the forgot-password link below.
