https://github.com/nbdy/vidscaler

Video upscaler (cli/gui)
https://github.com/nbdy/vidscaler

gui real-esrgan upscaler video video-upscaling vidscaler

Last synced: 3 months ago
JSON representation

Video upscaler (cli/gui)

• Host: GitHub
• URL: https://github.com/nbdy/vidscaler
• Owner: nbdy
• License: mit
• Created: 2024-04-15T16:18:53.000Z (over 1 year ago)
• Default Branch: main
• Last Pushed: 2024-04-15T21:35:37.000Z (over 1 year ago)
• Last Synced: 2025-03-19T20:01:12.934Z (7 months ago)
• Topics: gui, real-esrgan, upscaler, video, video-upscaling, vidscaler
• Language: Python
• Homepage:
• Size: 39.1 KB
• Stars: 1
• Watchers: 1
• Forks: 0
• Open Issues: 0
• Metadata Files:
  • Readme: README.md
  • Changelog: CHANGELOG.md
  • License: LICENSE

Awesome Lists containing this project

README

          
# video-upscaler

[![](http://github-actions.40ants.com/nbdy/vidscaler/matrix.svg)](https://github.com/nbdy/vidscaler)

upscale (x2/x4/x8) a video or a directory of videos using the command line

## dependencies

- moviepy

- loguru

- opencv

- torch

- numpy

- tqdm

- [Real-ESRGAN](https://github.com/sberbank-ai/Real-ESRGAN)

## how to ..

### .. install

```shell

# from pypi

pip install vidscaler

# from git repo

pip install git+https://github.com/nbdy/vidscaler

```

### .. run

#### .. gui

![GUI](screenshot/gui.png)

```shell

vidscaler-gui

# or

vidscaler --gui

```

#### cli

```shell

vidscaler --help

usage: vidscaler [-h] -i INPUT -o OUTPUT [-c] [-s SCALE]

                   [-m MODEL_DIRECTORY] [-g] [-p]

options:

  -h, --help            show this help message and exit

  -i INPUT, --input INPUT

                        Input directory or file

  -o OUTPUT, --output OUTPUT

                        Output directory or file

  -c, --cpu             Use CPU mode

  -s SCALE, --scale SCALE

                        Upscaling factor

  -m MODEL_DIRECTORY, --model-directory MODEL_DIRECTORY

  -g, --gui             Run the GUI

  -p, --preview         Enable live preview

```

## todo

- [ ] pausing

- [ ] start from last stopped point

  - json {"path": "path to video file", "frame": frame} as checkpoint.json?

- [ ] total runtime estimation

  - get total frame count of all videos 

  - avg. each frame processing duration

  - avgT * frameN = duration

- [ ] confirmation box when exiting/stopping while upscaling is running