Neural Image Abstraction Using Long Smoothing B-Splines
This work provides a method for stylized vector graphics generation, which is incremental as it builds on existing differentiable vector graphics pipelines.
The authors tackled the problem of generating smooth and arbitrarily long paths in image-based deep learning systems by integrating smoothing B-splines into a differentiable vector graphics pipeline, enabling control over fidelity vs. simplicity tradeoffs and stylization in geometric and image spaces.
We integrate smoothing B-splines into a standard differentiable vector graphics (DiffVG) pipeline through linear mapping, and show how this can be used to generate smooth and arbitrarily long paths within image-based deep learning systems. We take advantage of derivative-based smoothing costs for parametric control of fidelity vs. simplicity tradeoffs, while also enabling stylization control in geometric and image spaces. The proposed pipeline is compatible with recent vector graphics generation and vectorization methods. We demonstrate the versatility of our approach with four applications aimed at the generation of stylized vector graphics: stylized space-filling path generation, stroke-based image abstraction, closed-area image abstraction, and stylized text generation.