LEARD-Net: Semantic segmentation for large-scale point cloud scene

Ziyin Zeng, Yongyang Xu, Zhong Xie, Wei Tang, Jie Wan, Weichao Wu

Abstract: Highlights•A local feature extraction module (LFE) to take advantage of color information.•A local feature aggregation module (LFA) to augment local information while capturing local significant features and entire local neighbor.•A ResiDense module to allow the network to focus on both local and comprehensive features.•Our method achieves excellent performance on large-scale benchmarks, including S3DIS, Toronto3D and Semantic3D.