让AI学设计师画矢量图，自动保持线条连续与对齐。
DesigNet: Learning to Draw Vector Graphics as Designers Do
- 用可微模块控制曲线连续性与线条对齐，模仿设计师操作。
- 在矢量图生成中实现更高精度的连续性和对齐性。
- 输出可直接编辑，适合专业设计流程使用。
近年来，基于AI的内容生成取得了显著进展。然而，神经网络与人类设计师的工作方式存在根本差异，导致二者协作困难。针对可缩放矢量图形（SVG），我们为神经网络引入设计师常用工具，如轴对齐和命令连接处的显式连续性控制。提出DesigNet，一种基于层次化Transformer-VAE的模型，直接处理带有连续命令参数化的SVG序列。主要贡献是两个可微模块：连续性自精炼模块，预测并强制执行每个曲线点的C⁰、G¹和C¹连续性，通过调整贝塞尔控制点实现；对齐自精炼模块，具备水平或垂直线的吸附功能。DesigNet生成可编辑轮廓，在连续性和对齐性上优于现有方法，显著提升输出质量，便于后续修改与专业设计流程集成。源代码：https://github.com/TomasGuija/DesigNet。
原文摘要 · Abstract (English)
AI-driven content generation has made remarkable progress in recent years. However, neural networks and human designers operate in fundamentally different ways, making collaboration between them challenging. We address this gap for Scalable Vector Graphics (SVG) by equipping neural networks with tools commonly used by designers, such as axis alignment and explicit continuity control at command junctions. We introduce DesigNet, a hierarchical Transformer-VAE that operates directly on SVG sequences with a continuous command parameterization. Our main contributions are two differentiable modules: a continuity self-refinement module that predicts $C^0$, $G^1$, and $C^1$ continuity for each curve point and enforces it by modifying Bézier control points, and an alignment self-refinement module with snapping capabilities for horizontal or vertical lines. DesigNet produces editable outlines and achieves competitive results against state-of-the-art methods, with notably higher accuracy in continuity and alignment. These properties ensure the outputs are easier to refine and integrate into professional design workflows. Source Code: https://github.com/TomasGuija/DesigNet.
Thank you to arXiv for use of its open access interoperability. PaperDance 不是 arXiv 官方产品；中文卡片由大模型生成，请以原文为准。