Hi Siddha,
Welcome to Julia. You are welcome to propose other approaches, but the
approach mentioned on the GSoC page is to adapt the existing Python
generators roughly as follows:
1. emit pure C wrappers (the current generator uses the CPython API)
2. emit Julia code for function calls to the C wrappers, object mappings,
etc.
A good place to start understanding the current generator is this answer:
http://answers.opencv.org/question/6618/how-python-api-is-generated/
and the existing generator code:
https://github.com/Itseez/opencv/tree/master/modules/python/src2
If you are just starting out with Julia, I would suggest to pick one OpenCV
filter and manually write C wrappers and the corresponding Julia code to
call same (along the lines of the templates used in gen2.py). This will be
a good way to get a feel for how the different pieces fit together - in
particular, Julia's C FFI, passing data arrays, and memory management.
Best,
Isaiah
On Mon, Mar 10, 2014 at 11:08 AM, Siddha Ganju <[email protected]>wrote:
> Hi everyone,
>
> With regards to the "Computer Visions using OpenCV" project suggested on
> GitHub, I can see that it includes immense Data Visualization techniques so
> that all the high computation can be represented graphically.
> This<http://julialang.org/images/ijulia.png> image
> reminded me of the output Wolfram Alpha provides. But I could not find any
> benchmark ratings of Wolfram and Julia. However, I have never used image
> recognition with it, so adding image recognition to Julia is a novel idea
> and I would love to start it off.
>
>
>
> Python wrappers exist for OpenCV, so to my understanding, I will be
> required to translate them so that an interface for Julia can be created.
> Can anyone assign me a bug to look into that is related to OpenCV. I am a
> novice to Julia, so please guide me.
>
>
>
> I have worked with OpenCV before, on a face recognition project at my
> college and also at the New York University Hackathon. I really want to be
> a part of this project.
>
>
> Thank you everyone for your time,
>
>
> Siddha
>
