[Bf-committers] development of conditional pass-thru node for compositor

Jianming jianming.tom at gmail.com
Thu Apr 21 08:08:39 CEST 2011


Thank you, LetterRip,

Firstly, I am sorry to reply email like this, cause I don't actually
know how to reply an email on mailing list with gmail website.
I don't know if I should turn off daily digest on the mailing list,
otherwise it seems I couldn't reply email with simple reply button.

>could you explain a bit more of your 'beginner' status - are you new
>to programming or just new to programming blender.

I am a beginner programmer and new to blender, this is the first FOSS
project and first software project for me.
But I have experience on compositing, computer vision and cg.

Actually I'm a student doing a FOSS course in CS school of ANU now. I
chose blender base on my background and motivation.
I have C background and know little python. I do CV staff in matlab.

>This summer we expect to have one or two students focused on
>matchmoving related coding (integration of libmv and improving libmv
>for the areas important to our usage).

I am interested in this, since it's a CV project which I knew.

>there are some nodes and sequencer plugins that have been developed
>for matting that haven't been integrated yet and could be ported to
>2.5x code base to use in the compositor.

Are these in svn server?

>The 'GIMP' has integrated SIOX and has a branch that has tools for
>getting a much cleaner plate (smarter feathering control).

>I know brecht did a SIOX patch a year or two ago and didn't feel it
>was worthwhile at that time, but I think that with the update to SIOX
>for l2009 Gimp GSOC project it might good enough to be useful.

These are digital image processing projects, I think SIOX has better result.
==========
Thank you so much for this email, you provided useful info, pointed
pathway for me.

I'll continue to research on blender.

cheers,

Jianming (Tom)


More information about the Bf-committers mailing list