Julia noob trying Computer Vision

machinevision

#1

Hi guys! I am new to Julia and am trying to implement some Computer Vision algorithms such as SIFT in Julia. Can you please suggest some resources and packages that might be helpful? Till now I have only gone through the code at https://learnxinyminutes.com/docs/julia/ … Thanks a lot.


#2

JuliaImages is the best place for handling image data in Julia.

The first thing I would do, though, is simply become comfortable with Julia itself. The manual is great.

If you want some more examples, the jupyter notebooks from the MIT “Numerical Computing in Julia” class are available online here: https://github.com/alanedelman/18.337_2017/tree/master/lectures

@tkoolen and I also recently ran a seminar introducing Julia. Our focus was on robotics applications, but the first few notebooks are quite general: https://github.com/rdeits/DynamicWalking2018.jl/tree/master/notebooks

There’s also http://learningjulia.com/ which has a few blog posts focused on computer vision from a Julia newcomer’s perspective. It hasn’t been updated in a while, but it should still be a useful resource.


#3

By the way, as far as I know, some of the computer vision tools in Julia have not yet been updated for the alpha release of Julia v0.7, so I would suggest using the stable v0.6.3 release of Julia if you want to try out those packages.

Work is under way now to update them, so this should be resolved soon:


#4

Thanks a lot Sir!!!


#5

Look at ImageFeatures.jl (https://juliaimages.github.io/ImageFeatures.jl/latest/index.html). It contains HOG, LBP, and a couple of other feature descriptor algorithms.


#6

Thanks for sharing the notebooks. In https://github.com/rdeits/DynamicWalking2018.jl/blob/master/notebooks/2.%20Julia%20is%20Fast.ipynb, your first link should (probably) point to https://github.com/stevengj/18S096/blob/master/lectures/lecture1/Boxes-and-registers.ipynb.


#7

Thanks, fixed!