2D Gaussian Splatting for Bézier Spline Line Art Vectorization
Tianhao Chen, Clara Fernandez-Labrador, Marteinn Oskarsson, Chuck Tappan, Olga Sorkine-Hornung, Markus Gross, Christopher Schroers, Abdelaziz Djelouah
Abstract
Vectorization of line art and sketches is an important problem in computer graphics and remains an active area of research. In this paper, we leverage several recent advances in the field to introduce a new state-of-the-art method for line art vectorization. First, we show that it is possible to go beyond heuristics to extract a meaningful set of strokes from a raster image. We leverage both depth prediction and semantic feature extraction models to propose a more principled approach to split the skeleton graph of the sketch into subgraphs to initialize a set of strokes, producing results that better align with artistic intent. Second, we model strokes as Bézier curves with both geometric and appearance components, and employ 2D Gaussian splatting for fast, differentiable rendering. This formulation enables efficient fitting of strokes to the input while jointly optimizing control points and brush texture. We further demonstrate promising results on video by incorporating temporal tracking and adaptive keyframe introduction. Our evaluation shows that we achieve state-of-the-art results for image reconstruction, while being fast and producing strokes of high quality. More importantly, the fitting is highly compatible with user input, through the correction of the splines and the selection of optimization parameters.
Ask Lune
Know this paper in minutes, not an afternoon.
Lune can read every page of it and the top-tier papers that cite it, and gives you its claims, its evidence and its limits, each one cited to the paper it comes from. No skimming, no AI slop.
Lineage
- CLIPasso: semantically-aware object sketching cites General virtual sketching framework for vector line art
- DeepSVG: A Hierarchical Generative Network for Vector Graphics Animation cites A Learned Representation for Scalable Vector Graphics
Builds on26
- Vision Transformers Need RegistersTimothée Darcet, Maxime Oquab, Julien Mairal, Piotr BojanowskiICLR 2024 · 769 citations
- DeepSVG: A Hierarchical Generative Network for Vector Graphics AnimationAlexandre Carlier, Martin Danelljan, Alexandre Alahi, Radu TimofteNeurIPS 2020 · 247 citations
- CLIPasso: semantically-aware object sketchingYael Vinker, Ehsan Pajouheshgar, Jessica Y. Bo, Roman Christian Bachmann et al.SIGGRAPH 2022 · 219 citations
- A Learned Representation for Scalable Vector GraphicsRaphael Gontijo Lopes, David Ha, Douglas Eck, Jonathon ShlensICCV 2019 · 153 citations
- General virtual sketching framework for vector line artHaoran Mo, Edgar Simo-Serra, Chengying Gao, Changqing Zou et al.SIGGRAPH 2021 · 59 citations
Related papers
- Differentiable Stroke Planning with Dual Parameterization for Efficient and High-Fidelity Painting CreationJinfan Liu, Wuze Zhang, Zhangli Hu, Zhehan Zhao et al.CVPR 2026
- End-to-End Line Drawing VectorizationHanyuan Liu, Chengze Li, Xueting Liu, Tien-Tsin WongAAAI 2022 · 11 citations
- Color by Numbers: Interactive Structuring and Vectorization of Sketch ImageryAmal Dev Parakkat, Marie-Paule Cani, Karan SinghCHI 2021 · 13 citations
- Bézier Splatting for Fast and Differentiable Vector Graphics RenderingXi Liu, Chaoyi Zhou, Nanxuan Zhao, Siyu HuangNeurIPS 2025 · 2 citations
- StrokeFusion: Vector Sketch Generation via Joint Stroke-UDF Encoding and Latent Sequence DiffusionJin Zhou, Yi Zhou, Hongliang Yang, Pengfei Xu et al.AAAI 2026 · 2 citations
