Antonios P. Sarikas

icon company Department of Chemistry, University of Crete icon location Crete, Greece icon location PhD student at Materials Modeling & Design Group.

Results 4 comments of Antonios P. Sarikas

Why is it necessary to add the identity matrix in the transeform.py

@pmorerio This is also what I am thinking. I was confused with the term `initialized` since this matrix is always initialized/added at each step (both at training and inference).

Training for variable number of points.

> Additionally, PointNet is not supposed to be affected by redundant points due to its max-pooling operation. It is therefore possible to just sample point clouds with replacement, to a...

Variable number of points for each pointcloud during training

> HI @brbrhpt > > As there are batch norm layers, using batch size = 1 would cause some issues. At training time, TensorFlow does require a fixed tensor size,...

Each cloud is zero-mean and normalized into an unit sphere.How is this achieved?

@Hero7749 What if the size of the point cloud matters for the task at hand? Should normalization be applied in this case?