Siddhartha Gairola

Mr. Paarth Neekhara, your time for whole dataset is not doing what it should. :P

Training is very slow

2

The training is very slow. Even the evaluate part takes hours to finish. How much time did it take to train the model?

Adding functionality to compute principal curvatures.

1

Computation of principal curvatures of a zernike surface and also accounting for normalizing factor, to normalize dx, dy to lay in a unit circle and then re-normalize. Refer: https://wp.optics.arizona.edu/visualopticslab/wp-content/uploads/sites/52/2016/08/Class12_08.pdf for...

In the CheckSharpness() function in file: https://github.com/cjpark87/rdt-scan/blob/master/app/src/main/java/edu/washington/cs/ubicomplab/rdt_reader/core/ImageProcessor.java, the SHARPNESS_THRESHOLD=0.8, but the check conditions makes it `(1-0.8) = 0.2` which implies that the sharpness of the test image should be only...

No Online Mining

1

Hi! Great effort on the implementation. One short coming for this could be that currently this uses Offline Mining. We could also make the training process more robust by adding...

Please could you mention the version of CGAL required?

In file fit_rotations_mex.cpp #endif should be outside parenthesis

1

In file fit_rotations_mex.cpp #endif should be outside parenthesis. Refer line 127 in fit_rotations_mex.cpp

Running the mex files in Scilab

4

Hi, what changes will be required to port this to Scilab? I mean the changes to be made to the cmake for the mex files.

Hi! Great work on the project and implementation. Would be grateful, if you could please let me know how the calibration images were generated? or what software was used to...