从程序化3D程序中学习3D表示形式

Learning 3D Representations from Procedural 3D Programs

November 25, 2024

作者: Xuweiyi Chen, Zezhou Cheng

cs.AI

摘要

自监督学习已成为从未标记的3D点云中获取可转移的3D表示的一种有前途的方法。与广泛可获得的2D图像不同，获取3D资产需要专业知识或专业的3D扫描设备，这使得难以扩展并引发版权担忧。为了解决这些挑战，我们提出了从程序化3D程序中学习3D表示的方法，这些程序可以自动生成使用简单基元和增强生成的3D形状。 值得注意的是，尽管缺乏语义内容，从这种合成数据集中学习到的3D表示在各种下游3D任务中表现出色，与从语义可识别的3D模型（例如飞机）中学到的最先进表示相当，包括形状分类、部分分割和遮罩点云完成。我们的分析进一步表明，当前的自监督学习方法主要捕捉几何结构而不是高级语义。

English

Self-supervised learning has emerged as a promising approach for acquiring transferable 3D representations from unlabeled 3D point clouds. Unlike 2D images, which are widely accessible, acquiring 3D assets requires specialized expertise or professional 3D scanning equipment, making it difficult to scale and raising copyright concerns. To address these challenges, we propose learning 3D representations from procedural 3D programs that automatically generate 3D shapes using simple primitives and augmentations. Remarkably, despite lacking semantic content, the 3D representations learned from this synthesized dataset perform on par with state-of-the-art representations learned from semantically recognizable 3D models (e.g., airplanes) across various downstream 3D tasks, including shape classification, part segmentation, and masked point cloud completion. Our analysis further suggests that current self-supervised learning methods primarily capture geometric structures rather than high-level semantics.