Differentiable Stroke Planning with Dual Parameterization for Efficient and High-Fidelity Painting Creation
Jinfan Liu, Wuze Zhang, Zhangli Hu, Zhehan Zhao, Ye Chen, Bingbing Ni
Abstract
In stroke-based rendering, search methods often get trapped in local minima due to discrete stroke placement, while differentiable optimizers lack structural awareness and produce unstructured layouts. To bridge this gap, we propose a dual representation that couples discrete polylines with continuous Bézier control points via a bidirectional mapping mechanism. This enables collaborative optimization: local gradients refine global stroke structures, while content-aware stroke proposals help escape poor local optima. Our representation further supports Gaussian-splatting-inspired initialization, enabling highly parallel stroke optimization across the image. Experiments show that our approach reduces the number of strokes by 30-50%, achieves more structurally coherent layouts, and improves reconstruction quality, while cutting optimization time by 30-40% compared to existing differentiable vectorization methods.
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
- Stroke-based Neural Painting and Stylization with Dynamically Predicted Painting Region cites Learning to Paint With Model-Based Deep Reinforcement Learning
- Stroke-based Neural Painting and Stylization with Dynamically Predicted Painting Region cites Paint Transformer: Feed Forward Neural Painting with Stroke Prediction
- Depth Anything V2 cites 3D Gaussian Splatting for Real-Time Radiance Field Rendering
- Paint Transformer: Feed Forward Neural Painting with Stroke Prediction cites Learning to Paint With Model-Based Deep Reinforcement Learning
Builds on11
- 3D Gaussian Splatting for Real-Time Radiance Field RenderingBernhard Kerbl, Georgios Kopanas, Thomas Leimkühler, George DrettakisSIGGRAPH 2023 · 5,687 citations
- Depth Anything V2Lihe Yang, Bingyi Kang, Zilong Huang, Zhen Zhao et al.NeurIPS 2024 · 2,305 citations
- Learning to Paint With Model-Based Deep Reinforcement LearningZhewei Huang, Shuchang Zhou, Wen HengICCV 2019 · 180 citations
- Paint Transformer: Feed Forward Neural Painting with Stroke PredictionSonghua Liu, Tianwei Lin, Dongliang He, Fu Li et al.ICCV 2021 · 106 citations
- Stroke-based Neural Painting and Stylization with Dynamically Predicted Painting RegionTeng Hu, Ran Yi, Haokun Zhu, Liang Liu et al.ACM MM 2023 · 23 citations
Related papers
- Bézier Splatting for Fast and Differentiable Vector Graphics RenderingXi Liu, Chaoyi Zhou, Nanxuan Zhao, Siyu HuangNeurIPS 2025 · 2 citations
- 2D Gaussian Splatting for Bézier Spline Line Art VectorizationTianhao Chen, Clara Fernandez-Labrador, Marteinn Oskarsson, Chuck Tappan et al.SIGGRAPH 2026
- 2D Triangle Splatting for Direct Differentiable Mesh TrainingKaifeng Sheng, Zheng Zhou, Yingliang Peng, Qianwei WangCVPR Findings 2026
- Rethinking Style Transfer: From Pixels to Parameterized BrushstrokesDmytro Kotovenko, Matthias Wright, Arthur Heimbrecht, Björn OmmerCVPR 2021
- BG-Triangle: Bezier Gaussian Triangle for 3D Vectorization and RenderingMinye Wu, Haizhao Dai, Kaixin Yao, Tinne Tuytelaars et al.CVPR 2025
