论文标题
拉迪斯：3D形状编辑的语言删除
LADIS: Language Disentanglement for 3D Shape Editing
论文作者
论文摘要
自然语言互动是使3D形状设计民主化的有希望的方向。但是，现有的文本驱动3D形状编辑方法在产生脱钩的本地编辑中面临挑战为3D形状。我们通过学习在3D几何形状中的基础语言中学习的潜在表示来解决这个问题。为此，我们提出了一个补充工具集，包括新型网络架构，分离损失和新的编辑程序。此外，为了衡量编辑局部性，我们定义了一个新的指标，我们称之为部分编辑精度。我们表明，我们的方法在编辑区域方面优于现有的SOTA方法，而语言参考分辨率精度最高为6.6％。我们的工作表明，通过仅删除语言表示，下游3D形状的编辑也可以成为相关部分的本地化，即使该模型从未得到明确的基于零件的监督。
Natural language interaction is a promising direction for democratizing 3D shape design. However, existing methods for text-driven 3D shape editing face challenges in producing decoupled, local edits to 3D shapes. We address this problem by learning disentangled latent representations that ground language in 3D geometry. To this end, we propose a complementary tool set including a novel network architecture, a disentanglement loss, and a new editing procedure. Additionally, to measure edit locality, we define a new metric that we call part-wise edit precision. We show that our method outperforms existing SOTA methods by 20% in terms of edit locality, and up to 6.6% in terms of language reference resolution accuracy. Our work suggests that by solely disentangling language representations, downstream 3D shape editing can become more local to relevant parts, even if the model was never given explicit part-based supervision.
