Corey J. Nolet

Watershed function not implemented in image package

1

I've tried to run your algorithm now in both Octave 3.8 and 4.0 with their respect image packages. Both of these distributions tell me that, though the "watershed" function belongs...

Some bounding boxes are twisted in a way where they don't properly represent the text

2

In a few cases, it almost appears the bottom left and the top right boxes are flipped causing the bounding box to almost appear as two triangles that are touching...

The rotations of some word bounding boxes do not line up properly with the word itself.

4

It's strange and it happens quite often, maybe 20% of the time. It usually occurs when a word is rotated itself or has been projected onto a surface causing it...

The height of bounding boxes for some words is much larger than the words

9

This is causing some problems when determining the midpoints for the bounding boxes for the fully convolutional network. It's basically putting the midpoint in a different spot then it would...

I have some images that have quite a bit of space open for text to be drawn but only a single 2-3 character word will be drawn on the entire...

Have you tried any type of stochastic pooling for regularization similar to dropout?

5

I'm training my network now and from epoch to epoch, I'm testing the model to see how well it is in predictions. So far, it's getting me less than 25%...

GPU Accelerate Annoy Index Building / Queries

2

We have seen much speedup by executing both the random projection and tree building computations in GPUs. I am thinking the querying might also benefit from performance gains on GPUs....

`cupyx.scipy.spatial` and `cupyx.scipy.spatial.distance` packages

6

### Description The RAFT project (https://github.com/rapidsai/raft) has several building blocks for different algorithms in the RAPIDS Analytics ecosystem and among these building blocks are spatial distance and nearest neighbors algorithms....