PointNeXt: Revisiting PointNet++ with Improved Training and Scaling Strategies

Abstract

PointNet++ is one of the most influential neural architectures for point cloud understanding. Although the accuracy of PointNet++ has been largely surpassed by recent networks such as PointMLP and Point Transformer, we find that a large portion of the performance gain is due to improved training strategies, i.e. data augmentation and optimization techniques, and increased model sizes rather than architectural innovations. Thus, the full potential of PointNet++ has yet to be explored. In this work, we revisit the classical PointNet++ through a systematic study of model training and scaling strategies, and offer two major contributions. First, we propose a set of improved training strategies that significantly improve PointNet++ performance. For example, we show that, without any change in architecture, the overall accuracy (OA) of PointNet++ on ScanObjectNN object classification can be raised from 77.9% to 86.1%, even outperforming state-of-the-art PointMLP. Second, we introduce an inverted residual bottleneck design and separable MLPs into PointNet++ to enable efficient and effective model scaling and propose PointNeXt, the next version of PointNets. PointNeXt can be flexibly scaled up and outperforms state-of-the-art methods on both 3D classification and segmentation tasks. For classification, PointNeXt reaches an overall accuracy of 87.7 on ScanObjectNN, surpassing PointMLP by 2.3%, while being 10x faster in inference. For semantic segmentation, PointNeXt establishes a new state-of-the-art performance with 74.9% mean IoU on S3DIS (6-fold cross-validation), being superior to the recent Point Transformer. The code and models are available at https://github.com/guochengqian/pointnext.

Accepted by NeurIPS'22. Code and models are available at https://github.com/guochengqian/pointnext

Cited by in corpus (14)

• Deep Learning-based 3D Point Cloud Classification: A Systematic Survey and Outlook
• Point Cloud Classification Using Content-based Transformer via Clustering in Feature Space
• Advancements in Point Cloud-Based 3D Defect Detection and Classification for Industrial Systems: A Comprehensive Survey
• Regress Before Construct: Regress Autoencoder for Point Cloud Self-supervised Learning
• Beyond First Impressions: Integrating Joint Multi-modal Cues for Comprehensive 3D Representation
• Large Language Models and 3D Vision for Intelligent Robotic Perception and Autonomy
• On-the-fly Point Feature Representation for Point Clouds Analysis
• From CAD models to soft point cloud labels: An automatic annotation pipeline for cheaply supervised 3D semantic segmentation
• PosDiffNet: Positional Neural Diffusion for Point Cloud Registration in a Large Field of View with Perturbations
• Adaptive Margin Contrastive Learning for Ambiguity-aware 3D Semantic Segmentation
• PolarNet: 3D Point Clouds for Language-Guided Robotic Manipulation
• RESSCAL3D: Resolution Scalable 3D Semantic Segmentation of Point Clouds
• Learning Localization of Body and Finger Animation Skeleton Joints on Three-Dimensional Models of Human Bodies
• Scalable 3D Panoptic Segmentation As Superpoint Graph Clustering