PointNet and PointNet++ implemented by pytorch (pure python) and on ModelNet, ShapeNet and S3DIS.

PointCNN: Convolution On X-Transformed Points (NeurIPS 2018)

[NeurIPS 2019, Spotlight] Point-Voxel CNN for Efficient 3D Deep Learning

The official implementation of "Pix2Vox: Context-aware 3D Reconstruction from Single and Multi-view Images". (Xie et al., ICCV 2019)

Pixel2Mesh++: Multi-View 3D Mesh Generation via Deformation. In ICCV2019.

We evaluate our method on different datasets (including ShapeNet, CUB-200-2011, and Pascal3D+) and achieve state-of-the-art results, outperforming all the other supervised and unsupervised methods and 3D representations, all in terms of performance, accuracy, and training time.

(latest updates and bug fixed) DISN: Deep Implicit Surface Network for High-quality Single-view 3D Reconstruction

code for "Neural Cages for Detail-Preserving 3D Deformations"

Torch Implementation of NIPS'16 paper: Perspective Transformer Nets

Source code for: Flex-Convolution (Million-Scale Point-Cloud Learning Beyond Grid-Worlds), accepted at ACCV 2018

PVT: Point-Voxel Transformer for 3D Deep Learning

Rendering color and depth images for ShapeNet models.

Official implementation of GraphX-Convolution

Graph-convolutional GAN for point cloud generation. Code from ICLR 2019 paper Learning Localized Generative Models for 3D Point Clouds via Graph Convolution

ShapeGlot: Learning Language for Shape Differentiation

Supplementary code for SHREC 2017 RGB-D Object-to-CAD Retrieval track

Learning Graph-Convolutional Representations for Point Cloud Denoising (ECCV 2020)

Given an image and a target viewpoint, generate synthetic image in the target viewpoint

PRIN/SPRIN: On Extracting Point-wise Rotation Invariant Features

PyTorch implementation to train MortonNet and use it to compute point features. MortonNet is trained in a self-supervised fashion, and the features can be used for general tasks like part or semantic segmentation of point clouds.