TL;DR
This paper introduces a novel method for applying artistic style transfer to videos, ensuring visual consistency and stability across frames by leveraging advanced initialization and loss functions.
Contribution
It presents a new approach that extends image style transfer techniques to videos, addressing challenges like motion and occlusion for the first time.
Findings
Produces visually consistent stylized videos
Outperforms baseline methods in quality and stability
Handles large motion and occlusion effectively
Abstract
In the past, manually re-drawing an image in a certain artistic style required a professional artist and a long time. Doing this for a video sequence single-handed was beyond imagination. Nowadays computers provide new possibilities. We present an approach that transfers the style from one image (for example, a painting) to a whole video sequence. We make use of recent advances in style transfer in still images and propose new initializations and loss functions applicable to videos. This allows us to generate consistent and stable stylized video sequences, even in cases with large motion and strong occlusion. We show that the proposed method clearly outperforms simpler baselines both qualitatively and quantitatively.
Peer Reviews
No public reviews on file for this paper yet. If you reviewed it on a platform where reviews are public (OpenReview, ICLR, NeurIPS, ICML), you can paste yours below so the community can read it here.
Code & Models
Videos
Artistic Style Transfer For Videos | Two Minute Papers #68· youtube
