search

charlesq34 / pointnet

PointNet: Deep Learning on Point Sets for 3D Classification and Segmentation
Other
4.71k stars 1.44k forks source link

point cloud visualization #116

Open kargarisaac opened 6 years ago

kargarisaac commented 6 years ago

I use vispy to visualize the dataset. I want to visualize semantic segmentation dataset. You mentioned in paper that we have 9 numbers for [x,y,z,r,g,b,xn,yn,zn] for 13 classes, right? But when I see the numbers, there are more than 13 different rgb colors. I completely confused.

Would you please explain about the data and how we can visualize it? which library? is there any code? I want to see something like you plot in the picture in github repo.

kargarisaac commented 6 years ago

I solved the problem: My mistake: rgb in 9 features is not the label. I think it is color information from rgb image. I used vispy library and label from dataset as follows:

image

yiakwy commented 5 years ago

Good. But I found my app.run() does not work inside jupyter even I use webgl backend. No error (I compiled from source) but just no interactive js windows.

yiakwy commented 5 years ago

vispy does not work for the moment. See https://github.com/vispy/vispy/issues/1612, I resolve the problem using pythreejs.

Not you should insall vispy from source.