search

charlesq34 / pointnet2

PointNet++: Deep Hierarchical Feature Learning on Point Sets in a Metric Space
Other
3.02k stars 888 forks source link

LiDAR Point Cloud Classification results not good with real data #221

Open manishmaruthi opened 3 years ago

manishmaruthi commented 3 years ago

Dear all,

I am using DGCNN to classify LiDAR pointClouds. I have trained the model using ModelNet40 train data(2048 XYZ points, 250 epochs) and results are good when I try to classify objects using ModelNet40 test data.

But when I try to classify real data collected by velodyne sensor the prediction is mostly wrong. Please find the attached example. Most of the times I get output as Plant, Guitar or Stairs. I have shifted my objects to center of the coordinate frame and have normalized the values[-1,1]. I have even tried to clean the boundaries.

Can somebody suggest me what I could be doing wrong? Chair_local Person_local

manishmaruthi commented 3 years ago

I found that I have to feed the point cloud in the right orientation or else result can be bad. The network is invariant to geometric transform like rotation only up to an extent