Dongsu Zhang, Changwoon Choi, Inbum Park, Young Min Kim
TL;DR该研究提出了一种面向大规模 3D 场景连续几何的概率形状完成方法,该方法利用生成细胞自动机学习多模态分布,通过稀疏体素嵌入渐进式生成连续形状,其训练目标最大化完整形状分布的变分下限,并通过实验证明了相比于确定性模型,该方法在几何完成方面的表现更佳。
Abstract
We propose a probabilistic shape completion method extended to the continuous
geometry of large-scale 3d scenes. Real-world scans of 3D scenes suffer from a
considerable amount of missing data cluttered with unse