Previous approaches to generate shapes in a 3D setting train a GAN on the latent space of an autoencoder (AE). Even though this produces convincing results, it has two major shortcomings. As the GAN is limited to reproduce the dataset the AE was trained on, we cannot reuse a trained AE for novel data. Furthermore, it is difficult to add spatial supervision into the generation process, as the AE only gives us a global representation. To remedy these issues, we propose to train the GAN on grids (i.e. each cell covers a part of a shape). In this representation each cell is equipped with a latent vector provided by an AE. This localized representation enables more expressiveness (since the cell-based latent vectors can be combined in novel ways) as well as spatial control of the generation process (e.g. via bounding boxes). Our method outperforms the current state of the art on all established evaluation measures, proposed for quantitatively evaluating the generative capabilities of GANs. We show limitations of these measures and propose the adaptation of a robust criterion from statistical analysis as an alternative.

中文翻译：

---

使用基于网格的隐式函数生成 3D 形状

以前在 3D 设置中生成形状的方法在自动编码器 (AE) 的潜在空间上训练 GAN。尽管这产生了令人信服的结果，但它有两个主要缺点。由于 GAN 仅限于重现 AE 所训练的数据集，我们不能将经过训练的 AE 重用于新数据。此外，很难在生成过程中添加空间监督，因为 AE 只给我们一个全局表示。为了解决这些问题，我们建议在网格上训练 GAN（即每个单元格覆盖形状的一部分）。在这个表示中，每个细胞都配备了一个由 AE 提供的潜在向量。这种本地化表示能够实现更多的表达能力（因为可以以新颖的方式组合基于细胞的潜在向量）以及生成过程的空间控制（例如，通过边界框）。我们的方法在所有已建立的评估措施上都优于当前最先进的技术，建议用于定量评估 GAN 的生成能力。我们展示了这些措施的局限性，并建议采用统计分析中的稳健标准作为替代方案。