Scholarlink

<p>Semantic segmentation of point clouds is a fundamental task in environmental perception for autonomous driving. Due to the irregularity and sparsity of LiDAR data, efficiently processing large-scale unstructured point clouds remains a significant challenge. This paper presents LERPNet, a lightweight range-view-based semantic segmentation network. Specifically, we replace the Stem layer in FRNet with the proposed FD-Stem, which improves feature representation while reducing computational complexity. The proposed Adaptive Spatial-Channel Attention Enhancement module introduces long-range modeling capability with limited additional parameters, enabling effective learning of both spatial and channel-wise representations. Furthermore, the Range Point Nearest Neighbor Fusion Head alleviates the many-to-one mapping issue with lower inference latency, thereby improving segmentation performance. Extensive experiments demonstrate that LERPNet achieves competitive mIoU of 70.1% on SemanticKITTI, 79.2% on nuScenes, and 61.5% on SemanticSTF. Compared to recent state-of-the-art approaches, LERPNet operates at a real-time inference speed of 27 FPS with only 8.8M parameters. This outstanding trade-off between efficiency and accuracy highlights its practical advantages, making it highly suitable for real-time LiDAR semantic segmentation in resource-constrained autonomous driving scenarios.</p>

错误：搜索内容不能为空，请输入英文关键词
错误：关键词超出字数限制，请精简
高级检索

LERPNet: a lightweight and efficient range view point cloud semantic segmentation network

  • 期刊论文

  • 发表日期：

    2026年8月5日

查看全文
  • Yadong Guo,
  • Jing Liu,
  • Wei Zheng,
  • Wenling Xue,
  • Zijun Lin

摘要

Semantic segmentation of point clouds is a fundamental task in environmental perception for autonomous driving. Due to the irregularity and sparsity of LiDAR data, efficiently processing large-scale unstructured point clouds remains a significant challenge. This paper presents LERPNet, a lightweight range-view-based semantic segmentation network. Specifically, we replace the Stem layer in FRNet with the proposed FD-Stem, which improves feature representation while reducing computational complexity. The proposed Adaptive Spatial-Channel Attention Enhancement module introduces long-range modeling capability with limited additional parameters, enabling effective learning of both spatial and channel-wise representations. Furthermore, the Range Point Nearest Neighbor Fusion Head alleviates the many-to-one mapping issue with lower inference latency, thereby improving segmentation performance. Extensive experiments demonstrate that LERPNet achieves competitive mIoU of 70.1% on SemanticKITTI, 79.2% on nuScenes, and 61.5% on SemanticSTF. Compared to recent state-of-the-art approaches, LERPNet operates at a real-time inference speed of 27 FPS with only 8.8M parameters. This outstanding trade-off between efficiency and accuracy highlights its practical advantages, making it highly suitable for real-time LiDAR semantic segmentation in resource-constrained autonomous driving scenarios.