/pointer

Point TransformER - Paper Collection of Transformer based, Unsupervised and Self-supervised Point Cloud Understanding

Introduction

Since the Transformer architecture and self-supervised learning have witnessed the overwhelming applications in natural language processing, and recently, the vision community also embraces this trend. In 3D, actually there have been number of relevant attempts but lack of summary. Therefore, this repository provides a paper collection of point cloud processing focusing on the following 2 aspects:

  • unsupervised and self-supervised methods
  • Transformer-based models

Venue and Code are attached to each paper. Following the Paper link, you can also find its .bib file. We will supplement new paper regularly. If you find some related and important paper absent in this collection, feel free to raise a pull request, or contact Mr.sunhy@outlook.com.

Welcome your contributions! 😃

Datasets

Paper Venue Year Code
ModelNet - 3D ShapeNets: A Deep Representation for Volumetric Shapes CVPR 2015 link
ScanObjectNN - Revisiting Point Cloud Classification: A New Benchmark Dataset and Classification Model on Real-World Data ICCV 2019 link
ShapeNet: An Information-Rich 3D Model Repository ArXiv 2015 link
S3DIS: Joint 2D-3D-Semantic Data for Indoor Scene Understanding ArXiv 2017 link
ScanNet: Richly-Annotated 3D Reconstructions of Indoor Scenes CVPR 2017 link

Basics

Paper Venue Year Code
PointNet: Deep Learning on Point Sets for 3D Classification and Segmentation CVPR 2017 link
PointNet++: Deep Hierarchical Feature Learning on Point Sets in a Metric Space NeurIPS 2017 link
DGCNN: Dynamic Graph CNN for Learning on Point Clouds TOG 2019 link
KPConv: Flexible and Deformable Convolution for Point Clouds ICCV 2019 link
Walk in the Cloud: Learning Curves for Point Clouds Shape Analysis ICCV 2021 link
Rethinking Network Design and Local Geometry in Point Cloud: A Simple Residual MLP Framework ICLR 2022 link

Unsupervised and Self-supervised Learning for Point Cloud Understanding

Paper Venue Year Code
Point-Bert: Pre-training 3D Point Cloud Transformers with Masked Point Modeling CVPR 2022 link
CrossPoint: Self-Supervised Cross-Modal Contrastive Learning for 3D Point Cloud Understanding CVPR 2022 link
Contrastive Boundary Learning for Point Cloud Segmentation CVPR 2022 link
SegContrast: 3D Point Cloud Feature Representation Learning Through Self-Supervised Segment Discrimination RAL 2022 link
Exploring Data-Efficient 3D Scene Understanding with Contrastive Scene Contexts CVPR 2021 link
Self-Supervised Learning on 3D Point Clouds by Learning Discrete Generative Models CVPR 2021 link
PoinTr: Diverse Point Cloud Completion with Geometry-Aware Transformers ICCV 2021 link
Self-Supervised Pretraining of 3D Features on any Point-Cloud ICCV 2021 link
Unsupervised Point Cloud Pre-training via Occlusion Completion ICCV 2021 link
PointContrast: Unsupervised Pre-training for 3D Point Cloud Understanding ECCV 2020 link
Self-Supervised Few-Shot Learning on Point Clouds NeurIPS 2020 link
Self-Supervised Deep Learning on Point Clouds by Reconstructing Space NeurIPS 2019 link
FoldingNet: Point Cloud Auto-encoder via Deep Grid Deformation CVPR 2018 link
Learning a Probabilistic Latent Space of Object Shapes via 3D Generative-Adversarial Modeling NIPS 2016 link
Learning Representations and Generative Models for 3D Point Clouds ICML 2018 link
View inter-prediction GAN: unsupervised representation learning for 3D shapes by learning global shape memories to support local view predictions AAAI 2019 None

Transformer for Point Cloud Processing

Paper Venue Year Code
Embracing Single Stride 3D Object Detector with Sparse Transformer CVPR 2022 link
Fast Point Transformer CVPR 2022 link
PVT: Point-Voxel Transformer for Point Cloud Learning ArXiv 2022 link
An End-to-End Transformer Model for 3D Object Detection ICCV 2021 link
Point Transformer ICCV 2021 link
Voxel Transformer for 3D Object Detection ICCV 2021 link
3D Object Detection with Pointformer CVPR 2021 link
Improving 3D Object Detection with Channel-wise Transformer ICCV 2021 link
Group-Free 3D Object Detection via Transformers ICCV 2021 link
Cloud Transformers: A Universal Approach To Point Cloud Processing Tasks ICCV 2021 link
PCT: Point cloud transformer CVM 2021 link
Perceiver: General Perception with Iterative Attention ICML 2021 link
Perceiver IO: A General Architecture for Structured Inputs & Outputs ICLR 2022 link