https://github.com/soooda/linediff
This repo contains my code for my master's thesis - Generative In-between Line-drawing Interpolation
https://github.com/soooda/linediff
frame-interpolation pytorch sketch
Last synced: 2 months ago
JSON representation
This repo contains my code for my master's thesis - Generative In-between Line-drawing Interpolation
- Host: GitHub
- URL: https://github.com/soooda/linediff
- Owner: Soooda
- Created: 2025-07-13T07:31:55.000Z (11 months ago)
- Default Branch: main
- Last Pushed: 2025-07-14T14:44:32.000Z (11 months ago)
- Last Synced: 2025-07-14T14:54:20.208Z (11 months ago)
- Topics: frame-interpolation, pytorch, sketch
- Language: Python
- Homepage:
- Size: 96.9 MB
- Stars: 0
- Watchers: 0
- Forks: 0
- Open Issues: 0
-
Metadata Files:
- Readme: README.md
Awesome Lists containing this project
README
# Generative In-between Line-drawing Interpolation
## Environment
```bash
conda create -n LineDiff python=3.13
conda activate LineDiff
pip install -r requirements.txt
```
## Weights
You can download the pre-trained weights via this [URL](https://www.alipan.com/s/wqtDwMUMqaH). This is a self-extracting archive file made by [`makeself`](https://github.com/megastep/makeself). Download `weights.run` file into the project directory and run:
```bash
./weights.run --keep
```
You can get rid of `weights.run` file after unzipping the `weights` folder.
Not working?
Try `chmod u+x weights.run` to give the file the right permission.
## Train
Use `train.py` and `train_synthesis.py` to train and fine-tune on your data. The default dataloaders for AnimeRun and STD-12K are in `/data/` folder.
## Test
Run:
```bash
python test.py --frame1 --frame2 --output
```
## References
* [FlowDiffuser](https://github.com/LA30/FlowDiffuser)
* [GMFSS](https://github.com/98mxr/GMFSS_Fortuna)
* [AnimeRun](https://github.com/lisiyao21/AnimeRun)