论文标题
UCSG-NET-无监督的发现建设性固体几何树
UCSG-Net -- Unsupervised Discovering of Constructive Solid Geometry Tree
论文作者
论文摘要
签名距离场（SDF）是3D网格的突出隐式表示。基于此类表示的方法实现了最先进的3D形状重建质量。但是，这些方法难以重建非凸形形状。一种补救措施是合并一个建设性的实体几何框架（CSG），该框架表示形状是分解为原始物的。它允许体现具有布尔操作的简单树表示的高复杂性和非凸度的3D形状。然而，现有的方法受到监督，需要在培训过程中预先给出的整个CSG解析树。相反，我们提出了一个模型，该模型在没有任何监督的情况下提取CSG解析树。我们的模型可以预测原语的参数，并通过可区分的指标函数对其SDF表示。通过发现布尔操作员树的结构共同实现。该模型动态选择哪种算子组合超过原始词会导致高保真度的重建。我们在2D和3D自动编码任务上评估我们的方法。我们表明，预测的解析树表示是可以解释的，可以在CAD软件中使用。
Signed distance field (SDF) is a prominent implicit representation of 3D meshes. Methods that are based on such representation achieved state-of-the-art 3D shape reconstruction quality. However, these methods struggle to reconstruct non-convex shapes. One remedy is to incorporate a constructive solid geometry framework (CSG) that represents a shape as a decomposition into primitives. It allows to embody a 3D shape of high complexity and non-convexity with a simple tree representation of Boolean operations. Nevertheless, existing approaches are supervised and require the entire CSG parse tree that is given upfront during the training process. On the contrary, we propose a model that extracts a CSG parse tree without any supervision - UCSG-Net. Our model predicts parameters of primitives and binarizes their SDF representation through differentiable indicator function. It is achieved jointly with discovering the structure of a Boolean operators tree. The model selects dynamically which operator combination over primitives leads to the reconstruction of high fidelity. We evaluate our method on 2D and 3D autoencoding tasks. We show that the predicted parse tree representation is interpretable and can be used in CAD software.
