Point-BERT: Pre-training 3D Point Cloud Transformers with Masked Point Modeling

Abstract

We present Point-BERT, a new paradigm for learning Transformers to generalize the concept of BERT to 3D point cloud. Inspired by BERT, we devise a Masked Point Modeling (MPM) task to pre-train point cloud Transformers. Specifically, we first divide a point cloud into several local point patches, and a point cloud Tokenizer with a discrete Variational AutoEncoder (dVAE) is designed to generate discrete point tokens containing meaningful local information. Then, we randomly mask out some patches of input point clouds and feed them into the backbone Transformers. The pre-training objective is to recover the original point tokens at the masked locations under the supervision of point tokens obtained by the Tokenizer. Extensive experiments demonstrate that the proposed BERT-style pre-training strategy significantly improves the performance of standard point cloud Transformers. Equipped with our pre-training strategy, we show that a pure Transformer architecture attains 93.8% accuracy on ModelNet40 and 83.1% accuracy on the hardest setting of ScanObjectNN, surpassing carefully designed point cloud models with much fewer hand-made designs. We also demonstrate that the representations learned by Point-BERT transfer well to new tasks and domains, where our models largely advance the state-of-the-art of few-shot point cloud classification task. The code and pre-trained models are available at https://github.com/lulutang0608/Point-BERT

Accepted to CVPR 2022, Project page: https://point-bert.ivg-research.xyz

References in corpus (18)

• An Image is Worth 16x16 Words: Transformers for Image Recognition at Scale
• Learning Transferable Visual Models From Natural Language Supervision
• A Survey on Visual Transformer
• Transformers in Vision: A Survey
• Language Models are Few-Shot Learners
• Deformable DETR: Deformable Transformers for End-to-End Object Detection
• Zero-Shot Text-to-Image Generation
• BEiT: BERT Pre-Training of Image Transformers
• Revisiting Unreasonable Effectiveness of Data in Deep Learning Era
• iBOT: Image BERT Pre-Training with Online Tokenizer
• Masked Autoencoders Are Scalable Vision Learners
• Rethinking Semantic Segmentation from a Sequence-to-Sequence Perspective with Transformers
• Self-Supervised Learning with Swin Transformers
• Self-Supervised Few-Shot Learning on Point Clouds
• End-to-End Human Pose and Mesh Reconstruction with Transformers
• MaX-DeepLab: End-to-End Panoptic Segmentation with Mask Transformers
• PoinTr: Diverse Point Cloud Completion with Geometry-Aware Transformers
• Self-supervised Point Cloud Prediction Using 3D Spatio-temporal Convolutional Networks

Cited by in corpus (2)

• PointSmile: Point Self-supervised Learning via Curriculum Mutual Information
• RadarMOSEVE: A Spatial-Temporal Transformer Network for Radar-Only Moving Object Segmentation and Ego-Velocity Estimation