BriefGPT.xyz
Feb, 2019
多尺度稀疏卷积的多尺度亲和力用于三维实例分割
MASC: Multi-scale Affinity with Sparse Convolution for 3D Instance Segmentation
HTML
PDF
Chen Liu, Yasutaka Furukawa
TL;DR
该研究提出了一种基于稀疏卷积和点亲和力预测的3D实例分割新方法，通过简单而有效的聚类算法，在ScanNet基准测试上大幅优于现有的方法。
Abstract
We propose a new approach for
3d instance segmentation
based on
sparse convolution
and
point affinity prediction
, which indicates the like
→