Regional-to-Local Point-Voxel Transformer for Large-Scale Indoor 3D Point Cloud Semantic Segmentation

Semantic segmentation of large-scale indoor 3D point cloud scenes is crucial for scene understanding but faces challenges in effectively modeling long-range dependencies and multi-scale features. In this paper, we present RegionPVT, a novel Regional-to-Local Point-Voxel Transformer that synergistica...

Full description

Saved in:
Bibliographic Details
Published in:Remote sensing (Basel, Switzerland) Switzerland), 2023-10, Vol.15 (19), p.4832
Main Authors: Li, Shuai, Li, Hongjun
Format: Article
Language:English
Subjects:
Analysis
Computational linguistics
deep learning
Image segmentation
Language processing
Learning
multi-scale feature
Natural language interfaces
Neural networks
point cloud
regional-to-local
Scene analysis
self-attention
Semantic segmentation
Semantics
Three dimensional models
Transformers
Citations: Items that this one cites
Items that cite this one
Online Access:Get full text
Tags: Add Tag
No Tags, Be the first to tag this record!
Description
Summary:Semantic segmentation of large-scale indoor 3D point cloud scenes is crucial for scene understanding but faces challenges in effectively modeling long-range dependencies and multi-scale features. In this paper, we present RegionPVT, a novel Regional-to-Local Point-Voxel Transformer that synergistically integrates voxel-based regional self-attention and window-based point-voxel self-attention for concurrent coarse-grained and fine-grained feature learning. The voxel-based regional branch focuses on capturing regional context and facilitating inter-window communication. The window-based point-voxel branch concentrates on local feature learning while integrating voxel-level information within each window. This unique design enables the model to jointly extract local details and regional structures efficiently and provides an effective and efficient solution for multi-scale feature fusion and a comprehensive understanding of 3D point clouds. Extensive experiments on S3DIS and ScanNet v2 datasets demonstrate that our RegionPVT achieves competitive or superior performance compared with state-of-the-art approaches, attaining mIoUs of 71.0% and 73.9% respectively, with significantly lower memory footprint.
ISSN:2072-4292
2072-4292
DOI:10.3390/rs15194832