charlesq34/pointnet

Isn't this section incorrect?

mathephysicist opened this issue · 1 comments

mathephysicist commented

pointnet/models/transform_nets.py

Line 83 in 539db60

with tf.variable_scope('transform_feat') as sc:

You set the weights and bias to zero. Then matmul.

Hence you will always have zero updates to the TNet that is prior.

jaidonlybbert commented

I believe you are correct. Also, the associated paper makes a leap here in the described output of the TNet. The (first) TNet should output a 3x3, however the final specified layer has an output dimension of 256. Unfortunate bit of ambiguity.

image