PointNet: Deep Learning on Point Sets for 3D Classification and Segmentation

• 类型：arxiv
• 标识：1612.00593
• 链接：https://arxiv.org/abs/1612.00593
• 主题：evaluation
• 主分类：multimodal
• 形态：method
• 被引：18312
• 被引来源：Semantic Scholar
• S2被引：18312
• OpenAlex被引：2884
• 影响力被引：3302
• TLDR：This paper designs a novel type of neural network that directly consumes point clouds, which well respects the permutation invariance of points in the input and provides a unified architecture for applications ranging from object classification, part segmentation, to scene semantic parsing.
• OpenAlex ID：W2950642167
• OpenAlex DOI：10.48550/arxiv.1612.00593
• DOI：10.48550/arxiv.1612.00593
• DOI来源：OpenAlex
• 开放获取：green
• 开放获取链接：https://arxiv.org/pdf/1612.00593
• OpenAlex更新：2026-08-25
• 待LLM分类：否
• 成熟度：production
• 场景：3D point cloud、classification、segmentation
• 标题中文：PointNet: Deep Learning on Point Sets for 3D Classification and Segmentation
• TLDR中文：本文设计了一种直接处理点云的新型神经网络，较好地尊重了输入点的置换不变性，并为从物体分类、部件分割到场景语义解析等应用提供了统一架构。
• 来源文件：
• [OpenAlex discover]
• [S2 enrich]