Yochengliu/Relation-Shape-CNN
PyTorch implementation of a CNN for 3D point cloud analysis achieving shape recognition, segmentation, and normal estimation.

Velocity · 7d
+0.2
★ / day
Trend
→steady
star history
This repository provides the official PyTorch implementation of Relation-Shape CNN, a deep learning method for processing 3D point clouds. The paper introduces relation-shape convolution to learn geometric features from unordered point sets. It addresses tasks including 3D shape classification, part segmentation, and normal estimation, and was published at CVPR 2019 as an Oral presentation.