Difference between revisions of "Point Cloud"

From
Jump to: navigation, search
Line 20: Line 20:
 
Although Convolutional Neural Networks are the state of the art techniques for 2D object detection, they do not perform well on 3D point cloud due to the sparse sensor data, therefore new techniques are needed. [http://medium.com/@SmartLabAI/3d-object-detection-from-lidar-data-with-deep-learning-95f6d400399a 3D Object Detection from LiDAR Data with Deep Learning | SmartLab AI - Medium]
 
Although Convolutional Neural Networks are the state of the art techniques for 2D object detection, they do not perform well on 3D point cloud due to the sparse sensor data, therefore new techniques are needed. [http://medium.com/@SmartLabAI/3d-object-detection-from-lidar-data-with-deep-learning-95f6d400399a 3D Object Detection from LiDAR Data with Deep Learning | SmartLab AI - Medium]
  
http://upload.wikimedia.org/wikipedia/commons/thumb/4/4c/Point_cloud_torus.gif/220px-Point_cloud_torus.gif
+
<img src="http://upload.wikimedia.org/wikipedia/commons/thumb/4/4c/Point_cloud_torus.gif/220px-Point_cloud_torus.gif" width="400" height="400">
  
 
<youtube>KHYnC6t0boQ</youtube>
 
<youtube>KHYnC6t0boQ</youtube>
Line 32: Line 32:
 
<youtube>rKL7cmJP0bg</youtube>
 
<youtube>rKL7cmJP0bg</youtube>
  
http://news.developer.nvidia.com/wp-content/uploads/2018/06/splatnet.png
+
<img src="http://news.developer.nvidia.com/wp-content/uploads/2018/06/splatnet.png" width="1000" height="600">
  
 
== [http://github.com/charlesq34/pointnet PointNet]  ==
 
== [http://github.com/charlesq34/pointnet PointNet]  ==
Line 39: Line 39:
 
<youtube>Cge-hot0Oc0</youtube>
 
<youtube>Cge-hot0Oc0</youtube>
  
http://www-cdn.qwertee.io/media/uploads/blog/deep_learning_with_point_clouds/pointnet_architecture.png
 
  
http://stanford.edu/~rqi/pointnet2/images/pnpp.jpg
+
<img src="http://www-cdn.qwertee.io/media/uploads/blog/deep_learning_with_point_clouds/pointnet_architecture.png" width="1000" height="700">
 +
 
 +
<img src="http://stanford.edu/~rqi/pointnet2/images/pnpp.jpg" width="1000" height="700">
 +
 
  
  

Revision as of 11:33, 3 November 2019

Youtube search... ...Google search

A point cloud is a set of data points in space. Point clouds are generally produced by 3D scanners, which measure a large number of points on the external surfaces of objects around them. As the output of 3D scanning processes, point clouds are used for many purposes, including to create 3D CAD models for manufactured parts, for metrology and quality inspection, and for a multitude of visualization, animation, rendering and mass customization applications. [A point cloud is a set of data points in space. Point clouds are generally produced by 3D scanners, which measure a large number of points on the external surfaces of objects around them. As the output of 3D scanning processes, point clouds are used for many purposes, including to create 3D CAD models for manufactured parts, for metrology and quality inspection, and for a multitude of visualization, animation, rendering and mass customization applications. Point Cloud and List of programs for Point Cloud processing | Wikipedia

Although Convolutional Neural Networks are the state of the art techniques for 2D object detection, they do not perform well on 3D point cloud due to the sparse sensor data, therefore new techniques are needed. 3D Object Detection from LiDAR Data with Deep Learning | SmartLab AI - Medium


SPLATNet

PointNet



Vote3Deep

SqueezeSeg