moviefeats_visual / README.md
alitourani's picture
[fix] - update README.md
ba9c47d
---
license: gpl-3.0
task_categories:
- feature-extraction
- image-classification
- video-classification
- image-feature-extraction
language:
- en
pretty_name: MovieFeats_Visual
size_categories:
- n>100G
---
# 🎬 SceneSense Dataset
The dataset contains visual features obtained from a wide range of movies (full-length), their shots, and free trailers. It can be used in recommendation, information retrieval, classification, _etc_. You can read more about it [on the GitHub repository](https://github.com/RecSys-lab/SceneSense).
<!-- [![arXiv](https://img.shields.io/badge/arXiv-2309.10461-b31b1b.svg)](https://arxiv.org/abs/2309.10461) -->
## πŸ“ƒ Table of Content
- [Dataset Stats](#stats)
## πŸ“Š Dataset Stats <a id="stats"></a>
You can find below the stats and characteristics of the dataset:
| Aspect | Value |
| ------------------------------------------- | --------- |
| **Total Number of Movies:** | 274 |
| **Average Frames per Movie:** | 7,732 |
| **Total Number of Frames/Feature Vectors:** | 2,118,647 |
Also, the stats when combined with **MovieLenz 25M** ([link](https://grouplens.org/datasets/movielens/25m/)) are listed below:
| Aspect | Value |
| ----------------------------------- | --------- |
| **Accumulative # of Genres:** | 723 |
| **Average Movie Ratings:** | 3.88/5 |
| **Total # of Users:** | 158,146 |
| **Accumulative # of Interactions:** | 2,869,024 |
In terms of the model-based characteristics of the datase:
| Data | Model | Total Files | Size on Disk |
| ----------- | ----- | ----------- | ------------- |
| Full Movies | incp3 | 84,872 | 35.8 GB |
| Full Movies | vgg19 | 84,872 | 46.1 GB |
| Movie Shots | incp3 | 16,713 | 7.01 GB |
| Movie Shots | vgg19 | 24,598 | 13.3 GB |
| Trailers | incp3 | 1,725 | 681 MB |
| Trailers | vgg19 | 1,725 | 885 MB |
| **Total** | - | **214,505** | **~103.8 GB** |