VoxelNet: End-to-End Learning for Point Cloud Based 3D Object Detection
点云投影到三维网格,每一个网格内通过随机取点加全连接传播,得到一个单一的feature-vector。关键是对于这样一个稀疏的4D矩阵,在后面如何快速进行矢量运算,这里使用稀疏表达,用hash table建立快速index。

From here you can search these documents. Enter your search terms below.
| Keys | Action |
|---|---|
| ? | Open this help |
| n | Next page |
| p | Previous page |
| s | Search |