论文标题
scan2part：对现实世界3D扫描的细粒度和分层零件级别的理解
Scan2Part: Fine-grained and Hierarchical Part-level Understanding of Real-World 3D Scans
论文作者
论文摘要
我们提出了Scan2Part，这是一种在现实世界中嘈杂的室内RGB-D扫描中细分对象各个部分的方法。为此，我们改变了室内场景中对象的部分层次结构，并探索它们对场景理解模型的影响。具体而言，我们使用基于U-NET的稀疏体系结构，该体系结构通过利用多尺度功能层次结构来捕获基础3D扫描几何几何的细节。为了训练我们的方法，我们介绍了SCAN2PART数据集，该数据集是第一个大规模集合，可在现实世界中的零件级别提供详细的语义标签。总共，我们提供242,081个对应关系，在2,477个塑形物体的53,618个零件零件和1,506个扫描仪场景之间，在两个空间分辨率为2 cm $ $^3 $和5 cm $^3 $中。作为输出，即使几何形状很粗或部分丢失，我们也能够预测细粒度的每个对象零件标签。
We propose Scan2Part, a method to segment individual parts of objects in real-world, noisy indoor RGB-D scans. To this end, we vary the part hierarchies of objects in indoor scenes and explore their effect on scene understanding models. Specifically, we use a sparse U-Net-based architecture that captures the fine-scale detail of the underlying 3D scan geometry by leveraging a multi-scale feature hierarchy. In order to train our method, we introduce the Scan2Part dataset, which is the first large-scale collection providing detailed semantic labels at the part level in the real-world setting. In total, we provide 242,081 correspondences between 53,618 PartNet parts of 2,477 ShapeNet objects and 1,506 ScanNet scenes, at two spatial resolutions of 2 cm$^3$ and 5 cm$^3$. As output, we are able to predict fine-grained per-object part labels, even when the geometry is coarse or partially missing.
