jiupinjia/stylized-neural-painting
PyTorch implementation of a CVPR 2021 paper that translates images into stylized paintings using a neural renderer and stroke parameter optimization.
Not currently ranked — collecting fresh signals.
star history
This repository provides a complete training and inference pipeline for an image-to-painting translation method. The approach generates vivid and realistic painting artworks by predicting a sequence of physically meaningful stroke parameters. It uses a differentiable neural renderer that imitates vector rendering, framing stroke prediction as a parameter-search process that maximizes similarity between input and rendered output. The method can be jointly optimized with neural style transfer to transfer visual style from reference images.
Frequently asked
- What is jiupinjia/stylized-neural-painting?
- PyTorch implementation of a CVPR 2021 paper that translates images into stylized paintings using a neural renderer and stroke parameter optimization.
- Is stylized-neural-painting open source?
- Yes — jiupinjia/stylized-neural-painting is open source, released under the CC0-1.0 license.
- What language is stylized-neural-painting written in?
- jiupinjia/stylized-neural-painting is primarily written in Python.
- How popular is stylized-neural-painting?
- jiupinjia/stylized-neural-painting has 1.6k stars on GitHub.
- Where can I find stylized-neural-painting?
- jiupinjia/stylized-neural-painting is on GitHub at https://github.com/jiupinjia/stylized-neural-painting.