顺序很重要：从序列化 VR 草图生成三维形状
Order Matters: 3D Shape Generation from Sequential VR Sketches
把笔画时序作为条件信号注入草图编码器，再驱动扩散式三维生成器，为「手绘/交互输入→三维内容」这类任务提供了可复用的顺序感知编码思路与合成数据自动生成管线。
原文依据 an order-aware sketch encoder coupled with a diffusion-based 3D generator
摘要译文
由模型翻译 · 术语统一 · 请对照原文核对关键结论VR 草图让用户能够直接在三维空间中探索和迭代想法，相比传统 CAD 软件更快、更直观。然而，现有的草图到形状模型忽略了笔画的时序顺序，丢弃了关于结构与设计意图的关键线索。我们提出 VRSketch2Shape，这是首个面向序列化 VR 草图的三维形状生成框架与多类别数据集。我们的贡献有三点：(i) 一条从任意形状自动生成有序 VR 草图的流水线；(ii) 一个包含四个类别、超过 2 万对合成草图-形状对与 900 对手绘草图-形状对的数据集；(iii) 一个顺序感知的草图编码器，并与基于扩散的三维生成器相结合。我们的方法比先前工作具有更高的几何保真度，并能在极少监督下从合成草图有效泛化到真实草图。所有数据与模型均已开源于 https://chenyizi086.github.io/VRSketch2Shape_website/。
英文原文摘要 980 字符
VR sketching lets users explore and iterate on ideas directly in 3D, offering a faster and more intuitive alternative to conventional CAD software. However, existing sketch-to-shape models ignore the temporal ordering of strokes, discarding crucial cues about structure and design intent. We introduce VRSketch2Shape, the first framework and multi-category dataset for 3D shape generation from sequential VR sketches. Our contributions are threefold: (i) an automated pipeline that generates ordered VR sketches from arbitrary shapes, (ii) a dataset comprising over 20k synthetic and 900 hand-drawn sketch-shape pairs across four categories, and (iii) an order-aware sketch encoder coupled with a diffusion-based 3D generator. Our approach yields higher geometric fidelity than prior work and generalizes effectively from synthetic to real sketches with minimal supervision. All data and models are released open-source on on https://chenyizi086.github.io/VRSketch2Shape_website/.