本文探讨了LiDAR数据的城市规模语义对齐的非深度学习方法及利用深度学习从低成本LiDAR和图像中预测密集点云的技术。通过表面法线作为中间表示，实现了稀疏点云到密集点云的转换。
1. Semantic Alignment of LiDAR Data at City Scale. Fisher Yu, Jianxiong Xiao, Thomas Funkhouser. CVPR 2015.
This paper proposes a new, non-DL approach other than ICP to align point clouds in LiDAR.
2. DeepLiDAR: Deep Surface Normal Guided Depth Prediction for Outdoor Scene from Sparse LiDAR Data and Single Color Image. CVPR 2019
This paper proposes a DL approach to estimate dense point clouds from sparse point clouds in a low-cost LiDAR and images (with image warping). It uses the surface normal as the intermediate representation to be learned.
扫一扫
9805
被折叠的 条评论
为什么被折叠?
到【灌水乐园】发言
