Window normalization: Enhancing point cloud understanding by unifying inconsistent point densities

Qi Wang

doi:10.1016/j.imavis.2025.105789

Window normalization: Enhancing point cloud understanding by unifying inconsistent point densities

2025-10-21

0

PRE

AI

DOI：10.1016/j.imavis.2025.105789

Original

Original request for help

Save

Abstract

En 中文

• PAGWN module unifies point densities and balances local-global features. • Boosts segmentation accuracy with clearer object boundaries. • Improves small and complex object recognition on S3DIS. • Achieves gains with minimal parameters and high efficiency. • Enhances robustness and generalizes well across datasets.

Window normalization: Enhancing point cloud understanding by unifying inconsistent point densities

Abstract

Journal

Organization