# From Trailers to Storylines: An Efficient Way to Learn from Movies

arXiv: https://arxiv.org/abs/1806.05341 · HTML: https://mlsanity.com/paper/1806.05341/

## Key facts

- Cited by 10 papers in the map (data through 14 Oct 2026).
- First posted on arXiv 14 Jun 2018.

## Built on by (most cited first, 10 of 10)

- [MovieCuts: A New Dataset and Benchmark for Cut Type Recognition](https://mlsanity.com/paper/2109.05569/) (2021)
- [Learning to Cut by Watching Movies](https://mlsanity.com/paper/2108.04294/) (2021)
- [Match Cutting: Finding Cuts with Smooth Visual Transitions](https://mlsanity.com/paper/2210.05766/) (2022)
- [Multi-modal Segment Assemblage Network for Ad Video Editing with Importance-Coherence Reward](https://mlsanity.com/paper/2209.12164/) (2022)
- [Transformation vs Tradition: Artificial General Intelligence (AGI) for Arts and Humanities](https://mlsanity.com/paper/2310.19626/) (2023)
- [Collaborative Noisy Label Cleaner: Learning Scene-aware Trailers for Multi-modal Highlight Detection in Movies](https://mlsanity.com/paper/2303.14768/) (2023)
- [InstaIndoor and Multi-modal Deep Learning for Indoor Scene Recognition](https://mlsanity.com/paper/2112.12409/) (2021)
- [Movie Genre Classification by Language Augmentation and Shot Sampling](https://mlsanity.com/paper/2203.13281/) (2022)
- [Find the Cliffhanger: Multi-Modal Trailerness in Soap Operas](https://mlsanity.com/paper/2401.16076/) (2024)
- [TAKE 85: Testing Audiovisual filmmaKer's intEnt across 85 Hours of Film](https://mlsanity.com/paper/2608.30068/) (2026)

## Abstract (arXiv)

The millions of movies produced in the human history are valuable resources for computer vision research. However, learning a vision model from movie data would meet with serious difficulties. A major obstacle is the computational cost -- the length of a movie is often over one hour, which is substantially longer than the short video clips that previous study mostly focuses on. In this paper, we explore an alternative approach to learning vision models from movies. Specifically, we consider a framework comprised of a visual module and a temporal analysis module. Unlike conventional learning methods, the proposed approach learns these modules from different sets of data -- the former from trailers while the latter from movies. This allows distinctive visual features to be learned within a reasonable budget while still preserving long-term temporal structures across an entire movie. We construct a large-scale dataset for this study and define a series of tasks on top. Experiments on this dataset showed that the proposed method can substantially reduce the training time while obtaining highly effective features and coherent temporal structures.