Vote3Deep: Fast Object Detection in 3D Point Clouds Using Efficient Convolutional Neural Networks
下图Vote3Deep在KITTI上不可见点云的检测效果。 本文方法：在原生的3D特征（native 3D representation）上执行一个稀疏卷积，接着是一个ReLU非线性，它返回一个新的稀疏3D特征。这个过程可以像传统CNN一样重复和叠加，输出层预测检测分数。用NMS对重复检测进行剪枝（3D中的NMS能够更好地处理彼此背后的对象，因为3D包围框的重叠小于它们的2D投影）。 本文为每个类
相关文章
- 1. Vote3Deep: Fast Object Detection in 3D Point Clouds Using Efficient CNN论文综述（2016）
- 2. PointPillars: Fast Encoders for 3D Object Detection from Point Clouds
- 3. Convolutional Neural Network - Object Detection
- 4. Fast Face-swap Using Convolutional Neural Networks
- 5. 【3D计算机视觉】Deep Hough Voting for 3D Object Detection in Point Clouds
- 6. PIXOR Real-time 3D Object Detection from Point Clouds
- 7. Convolutional Neural Networks（week 3）---Object detection
- 8. ShellNet:Efficient Point Cloud Convolutional Neural Networks using Concentric Shells Statics
- 9. 《论文阅读》ImVoteNet: Boosting 3D Object Detection in Point Clouds with Image Votes
- 10. Point Cloud 2019
- 更多相关文章...
- • SQL IN 操作符 - SQL 教程
- • Swift for-in 循环 - Swift 教程
- • RxJava操作符（六）Utility
- • 互联网组织的未来：剖析GitHub员工的任性之源
相关标签/搜索
每日一句
-
每一个你不满意的现在，都有一个你没有努力的曾经。
欢迎关注本站公众号,获取更多信息
相关文章
- 1. Vote3Deep: Fast Object Detection in 3D Point Clouds Using Efficient CNN论文综述（2016）
- 2. PointPillars: Fast Encoders for 3D Object Detection from Point Clouds
- 3. Convolutional Neural Network - Object Detection
- 4. Fast Face-swap Using Convolutional Neural Networks
- 5. 【3D计算机视觉】Deep Hough Voting for 3D Object Detection in Point Clouds
- 6. PIXOR Real-time 3D Object Detection from Point Clouds
- 7. Convolutional Neural Networks（week 3）---Object detection
- 8. ShellNet:Efficient Point Cloud Convolutional Neural Networks using Concentric Shells Statics
- 9. 《论文阅读》ImVoteNet: Boosting 3D Object Detection in Point Clouds with Image Votes
- 10. Point Cloud 2019