生成干净拓扑三维网格的缺失数据集是什么？
1 分•作者: fivepebbles•5 个月前
我们需要一个 Blender bpy.ops 数据集，用于自回归建模，而网格生成这条路正在被“不发表就灭亡”的压力吞噬，已经走到尽头了。
大家都知道神经网格生成有点像无用的玩具，但在 2022 年，Autodesk Research 发表了 SkexGen（ICML 2022），一个生成 CAD 构造序列的自回归模型。你绘制一个 2D 轮廓，对其进行拉伸、布尔运算，每一步都是有效的 CAD 操作。SCAD 和 Adam 等已经将其外推到 LLM，这很好——如果它们真的专业化，实际上会非常出色，而差距在于它和一个 LLM 视觉问题。
它之所以有效，是因为 CAD 文件原生存储了它们的构建历史，并且之所以存在，是因为构建序列始终被存储，而 DeepCAD 数据集免费为它们提供了数千个这样的序列。 Blender 的呢？
研究界选择了追求直接网格生成，将顶点和面标记成序列（PolyGen、MeshAnything、MeshXL、MESHTRON 等），或者逆向网格简化（ARMesh）。这些方法正在变得更好，但它们从根本上是在与表示作斗争。顶点坐标序列并没有编码边缘循环存在的原因，但 bpy.ops 却可以。我们应该创建一个数据集。 Blender 已经将每个 bpy.ops 调用记录到其信息面板中。一个录制插件可以捕获带有所有参数的完整 bpy.ops 调用、每一步的选择状态（选择了哪些顶点/边/面）、每一步（或关键间隔）的轻量级网格快照以及最终网格作为标签。
是的，噪声、上下文和尺度之间存在挑战，但是……？自回归方法已经得到验证。差距完全在于数据收集基础设施。为什么不呢？
查看原文
We need a Blender bpy.ops dataset for autoregressive modeling and mesh generation is a dead end being eaten alive by publish or perish.<p>Everyone knows neural mesh generation is a bit of a useless toy, but in 2022, Autodesk Research published SkexGen (ICML 2022), an autoregressive model that generates CAD construction sequences. You sketch a 2D profile, extrude it, boolean it, and each step is a valid CAD operation. SCAD and Adam and whatnot are already extrapolating that to LLMs, and it's fine - if they were actually specialized, they'd be quite good, actually, and the gap is that and an LLM vision problem.<p>And it works because CAD files natively store their construction history, and it exists because the construction sequences are always stored, and the DeepCAD dataset gave them thousands of these sequences for free. Where is that for Blender?<p>The research community has chosen to pursue direct mesh generation, tokenizing vertices and faces into sequences (PolyGen, MeshAnything, MeshXL, MESHTRON, etc.) or reversing mesh simplification (ARMesh). These approaches are getting better, but they're fundamentally fighting the representation. A sequence of vertex coordinates doesn't encode why an edge loop is there, but by.ops does. We should make a dataset. Blender already logs every bpy.ops call to its Info panel. A recording addon could capture the full bpy.ops call with all parameters, the selection state at each step (which vertices/edges/faces were selected), a lightweight mesh snapshot at each step (or at key intervals) and the final mesh as the label.<p>Yes, there's challenges between noise, context and scale, but...? The autoregressive method is proven already. The gap is entirely in data collection infrastructure. Why not?