Point clouds are characterized by irregularity and unstructuredness, which pose challenges in efficient data exploitation and discriminative feature extraction. In this paper, we present an unsupervised deep neural architecture called Flattening-Net to represent irregular 3D point clouds of arbitrary geometry and topology as a completely regular 2D point geometry image (PGI) structure, in which coordinates of spatial points are captured in colors of image pixels. \mr{Intuitively, Flattening-Net implicitly approximates a locally smooth 3D-to-2D surface flattening process while effectively preserving neighborhood consistency.} \mr{As a generic representation modality, PGI inherently encodes the intrinsic property of the underlying manifold structure and facilitates surface-style point feature aggregation.} To demonstrate its potential, we construct a unified learning framework directly operating on PGIs to achieve \mr{diverse types of high-level and low-level} downstream applications driven by specific task networks, including classification, segmentation, reconstruction, and upsampling. Extensive experiments demonstrate that our methods perform favorably against the current state-of-the-art competitors. We will make the code and data publicly available at https://github.com/keeganhk/Flattening-Net.
翻译:在本文中,我们展示了一个不受监督的深神经结构,称为Flattenning-Net,作为任意几何和地形学的非正常三维云,作为完全正常的 2D 点几何图象(PGI) 结构,其中空间点坐标以图像像素颜色显示。\mr{直观地显示,Flattenning-Net暗含地近似于一个本地平滑的3D至2D平坦进程,同时有效保持周边的一致性。}\mr{As通用代表模式,PGI内在地编码基本结构的固有属性,并促进地貌特征组合。}为了展示其潜力,我们直接在PGIS上建立一个统一的学习框架,以实现由特定任务网络驱动的下游应用,包括分类、分解、重建、加印。广泛的实验表明,我们的方法对当前州/州/州/州/州/ 网络可用的数据具有优势。