Automatic home video editing on music

Author(s): Weers, F.R.T. (2019)

Abstract:
A lot of videos are recorded during events (such as symposia), trips or other special occasions (like a wedding). Watching all the raw content is boring and takes a lot of time. Edited videos can summarize the experience in a fun and interesting way, but movie-editing still requires a lot of manual work and expertise. To solve this issue, we propose a method to automatically select important parts of photos and videos, analyze music for beats and use this to generate a short movie with transitions in rhythm. Automatic selection of important parts is based on motion detection, infidelity and simplicity. Combining the chosen song with the photos and videos is a constrained optimization problem and two solutions are given.

Document(s):

weers_BA_EEMCS.pdf