metadata
license: mit
language:
- en
tags:
- animation
- stroke
- interpolation
- 2D
- image
- video
Stroke Interpolation Model
To read the paper: https://drive.google.com/file/d/1EESd81NSs93OJYb42DartC5udTlOShRp/view?usp=sharing
Example
The model predicts the inbetween frames (mid frame), given two key frames.
Installation
pip install opencv-python
pip3 install torch torchvision torchaudio --index-url https://download.pytorch.org/whl/cu128
Run the Program
py combined.py
This will load the model from checkpoints.
Evaluate
py generate_eval.py
This generates images for evaluation.
py eval.py
This evaluates the generated images.
Test
py test.py
Add frames in test_frames folder.
py video.py
This combines those 3 frames into .mp4 format.
Dataset
Dataset is available at: Google Drive Link