Movie Summarization via Sparse Graph Construction

Pinelopi Papalampidi, Frank Keller, Mirella Lapata

Research output: Chapter in Book/Report/Conference proceedingConference contribution


We summarize full-length movies by creating shorter videos containing their most informative scenes. We explore the hypothesis that a summary can be created by assembling scenes which are turning points (TPs), i.e., key events in a movie that describe its storyline. We propose a model that identifies TP scenes by building a sparse movie graph that represents relations between scenes and is constructed using multimodal information. According to human judges, the summaries created by our approach are more informative and complete, and receive higher ratings, than the outputs of sequence-based models and general-purpose summarization algorithms. The induced graphs are interpretable, displaying different topology for different movie genres.
Original languageEnglish
Title of host publication2021 AAAI Conference on Artificial Intelligence
PublisherAAAI Press
Number of pages14
Publication statusAccepted/In press - 2 Dec 2020
EventThe Thirty-Fifth AAAI Conference on Artificial Intelligence - Virtual Conference
Duration: 2 Feb 20219 Feb 2021
Conference number: 35


ConferenceThe Thirty-Fifth AAAI Conference on Artificial Intelligence
Abbreviated titleAAAI-21
Internet address

Fingerprint Dive into the research topics of 'Movie Summarization via Sparse Graph Construction'. Together they form a unique fingerprint.

Cite this