Dmytro Mishkin

### Describe the bug https://huggingface.co/docs/datasets/en/semantic_segmentation shows wrong example with torchvision transforms. Specifically, as one can see in screenshot below, the object boundaries have weird colors. Original example with `albumentations` is...

Error while running an example from Readme

5

Hi, I ham running this code ``` init; % add project paths, add solver aliases etc. img_path = 'data/img1.jpg'; % specify a path to an image solvers = [Solver2PC4CA, Solver6CA];...

BatchNorm after ReLU

8

Hi, I am performing somehow similar benchmark, but on caffenet128 (and moving to ResNets now) on ImageNet. One thing, that I have found - the best position of BN in...

pretrained weights?

2

Hi, This seems as cool piece of work done :) Do you have somewhere pytorch weights of any of edge detectors?

Adds evaluation on challenging WxBS and EVD datasets

1

Closes https://github.com/cvg/glue-factory/issues/35 Compared to original papers, I have added the image resize to be consistent with gluefactory evals, and to make things easier for learned features. Also turned out, that...

Possibly faster training

4

This curves are with SuperPoint-Open I have been experimenting with faster training and it seems that OneCycle policy with `max_lr=3e-4` provides the same results (based on training and val metrics)...

Eval on WxBS

3

Hi, Would you be interested in PR, adding the evaluation on WxBS? https://github.com/ducha-aiki/wxbs-benchmark  And if yes - do you need it to be self-contained, or I just add `wxbs-benchmark`...

Hi, Just FYI, the configs are different compared to the paper: Paper: lr 1e-5, by 0.95 in each epoch after 10 epochs. > Training details: Weights are initialized from the...

Any numbers for new models (open SP, ALIKED, SIFT)?

3

Hi, I'd like to add some of this models to kornia if they are worth it. Given that you have trained them, I guess you have some numbers? If not,...