Dmytro Mishkin

[Do not merge]RANSAC improvements

2

#### Changes - Added LU solver for RANSAC-homography, it speeds-up the minimal-solver part x10 - Now find_fundamental weights is optional, as it was for homography - Fix test precision and...

## 🚀 Feature Integrate AdaLAM utils into core kornia, maybe reuse it - batch_2x2_inv - batch_2x2_det - Affine RANSAC

Implement classical object tracking algorithms like Lucas-Kande and KCF

3

## 🚀 Feature Since we have started our tracking module, it would be nice to populate it with classical and CNN-based things References: - http://robots.stanford.edu/cs223b04/algo_tracking.pdf - [Lukas-Kanade Tracker](https://www.ri.cmu.edu/pub_files/pub3/baker_simon_2002_3/baker_simon_2002_3.pdf) - [KCF](https://cw.fel.cvut.cz/b202/courses/mpv/labs/4b_tracking/start)

Hi, This is a great work! Would you consider adding your model to kornia? https://github.com/kornia/kornia I believe, that it could boost the usage and citations for you and make life...

Medium paywall

2

Hi, Is the medium paywall for your posts is intended or accidental?  Best, Dmytro

Datasets are not available

1

Hi, When I I trying to download datasets, it shows: ``` bash download.sh --2022-07-28 14:01:37-- https://docs.google.com/uc?export=download&confirm=&id=1ylw6AZ8AsyRmvook3FXrR2miSNajo6HE Resolving docs.google.com (docs.google.com)... 2a00:1450:4014:80e::200e, 142.251.36.142 Connecting to docs.google.com (docs.google.com)|2a00:1450:4014:80e::200e|:443... connected. HTTP request sent, awaiting...

Which model is really the best?

5

Hi, There are several models available. Which model is the best? I am on the way of integrating your models into https://github.com/kornia/kornia and would like to make good general purpose...

Simple script for keypoint extraction on user images?

4

Hi, Could please provide a script for detector usage? Like "detect and save into text file".

Number of parameters

1

Hi, If I understand source code correctly https://github.com/forestagostinelli/Learned-Activation-Functions-Source/blob/master/src/caffe/layers/apl_layer.cpp#L71 and https://github.com/forestagostinelli/Learned-Activation-Functions-Source/blob/master/src/caffe/layers/apl_layer.cpp#L30 then when used in convolution layers, APL learns not C_sums paramers, but H_W_C (== count/num) \* sums parameters. So same...

PR to kornia

1

Hi, I have created a PR to kornia https://github.com/kornia/kornia/pull/2436 Please tell me, if you have any feedback or objections. Best, Dmytro