Metadata-Version: 2.1
Name: vidfetch
Version: 0.0.1a1
Summary: data4co provides convenient dataset generators for the combinatorial optimization problem
Home-page: https://github.com/heatingma/VidFetch/
Author: heatingma
License: Mulan PSL v2
Classifier: License :: OSI Approved :: Mulan Permissive Software License v2 (MulanPSL-2.0)
Classifier: Programming Language :: Python
Classifier: Programming Language :: Python :: 3
Classifier: Programming Language :: Python :: 3.8
Classifier: Programming Language :: Python :: 3.9
Classifier: Programming Language :: Python :: 3.10
Classifier: Programming Language :: Python :: 3 :: Only
Classifier: Programming Language :: Python :: Implementation :: CPython
Classifier: Operating System :: Microsoft :: Windows
Classifier: Operating System :: POSIX
Classifier: Operating System :: Unix
Classifier: Operating System :: MacOS
Classifier: Environment :: Console
Classifier: Topic :: Scientific/Engineering :: Artificial Intelligence
Classifier: Topic :: Scientific/Engineering :: Image Recognition
Classifier: Topic :: Scientific/Engineering :: Mathematics
Requires-Python: >=3.8
Description-Content-Type: text/markdown
Requires-Dist: aiohttp>=3.9.3
Requires-Dist: async_timeout>=4.0.3
Requires-Dist: tqdm>=4.66.2
Requires-Dist: texttable>=1.7.0
Requires-Dist: moviepy>=1.0.3
Requires-Dist: bs4>=0.0.2
Requires-Dist: selenium>=4.18.1
Requires-Dist: requests>=2.31.0
Requires-Dist: texttable>=1.7.0
Requires-Dist: huggingface_hub>=0.22.2


<div align="center">
<img src="static/logo-2.png" alt="logo" width="750"/>
</div>


``Latte``, **a novel latent diffusion transformer for video generation**, utilizes spatio-temporal tokens extracted from input videos and employs a series of Transformer blocks to model the distribution of videos in the latent space. ``Latte`` achieves state-of-the-art performance on four standard video generation datasets ``FaceForensics``, ``SkyTimelapse``, ``UCF101``, and ``Taichi-HD``. [paper](https://arxiv.org/pdf/2401.03048v1.pdf), [code](https://github.com/Vchitect/Latte?tab=readme-ov-file), [pretrained](https://huggingface.co/maxin-cn/Latte)


However, ``Latte`` still falls short in terms of video generation length and quality compared to ``Sora``. To achieve training and generation effects close to Sora, the Latte model requires more high-quality text-video paired datasets. Therefore, we have created ``VidFetch``, an open-source dataset download tool to obtain copyright-free videos from various free video websites.


## Free Video Support

| website | windows | macos | linux |
| :-----: | :-----: | :---: | :---: |
| [Pexels](https://www.pexels.com) | ✔ | 📆 | 📆 |
| [Mazwai](https://mazwai.com/stock-video-footage) | 📆 | 📆 | 📆 |
| [Mixkit](https://mixkit.co/free-stock-video) | ✔ | 📆 | ✔ |
| [Pixabay](https://pixabay.com/videos/search/?order=ec) | ✔ | 📆 | 📆 |
| [Coverr](https://coverr.co/stock-video-footage) | 📆 | 📆 | 📆 |


## How to use

**Install related dependency packages**
```bash
pip install -r doc/requirements.txt
```


**You only need three lines of code to start downloading the video**
```python
from vidfetch.website import MixkitVideoDataset

mixkit = MixkitVideoDataset(root_dir="mixkit")
mixkit.download(platform="windows")
```
<details>
<summary>Click to view examples we have implemented</summary>

- Download videos from Mixkit

<div><img src="static/download.gif" width=100%></div>

- When you interrupt the download, the monitor will record the video information you downloaded successfully last time and continue downloading based on this information

<div><img src="static/continue_download.gif" width=100%></div>

</details>

## VidFetch's design philosophy
<img src="static/VidFetch.png" weight="100%">
