Introduction to Maya - Rendering in Arnold
This course will look at the fundamentals of rendering in Arnold. We'll go through the different light types available, cameras, shaders, Arnold's render settings and finally how to split an image into render passes (AOV's), before we then reassemble it i
# 1 16-02-2011 , 09:34 AM
Falott's Avatar
Registered User
Join Date: Jan 2005
Location: vienna
Posts: 1,095

Girlfriend to Avatar

Dear fellow community members

It´s about time to finally start a private project I had in mind for over a year now. Actually, yep – it´s going to be just another Avatar version of a girlfriend. But it´s my girl and what makes it that special for me is to show her something what I´ve been talking to her since we first met but could never really explain. I guess everyone who´s found their true love knows about that very special look from the partner, and if witnessed, makes you instantly fall in love again. This look but only happens as often as a butterfly would land on your toes and does not last for longer than the glimpse of an eye. I´ve never managed to capture this look with an external device to show her. Maybe it´s not even possible to make something like this visible with any scanning device, because it is a deeply inner reaction of oneself triggered through visual impression. Maybe this extra part can only be added by the witness himself to make it be seen. So, the advantage of us who are capable of managing 3D software is to not only recreate photo realism but create emotional expressions too – if we are good at what we do.

I thought it would be a good idea to share progress - not only for beginning people to get an interactive “from A to Z” tutorial but also for me to hear your input, criticism, and suggestions for improvement. So please anyone, don´t hesitate and feel invited if you like to ask questions or make comments of any kind. But beware! This project will possibly go on for ages... (hope I can get it finished within this year though)

And to say it right in the beginning I´ve never tackled a project like this. Although my personal interests always grounded on organic modelling and stuff, I´ve been used to make money mainly with architecture and product visuals so far.

To cut the long story short, here´s the plan:

Final image composition should be comprised of about 4 different views with different facial expressions in size of around 60*60 inches or something like that.

1)
modeling human bust in high resolution with as much detail within the geometry as needed. (folds, wrinkles) normalMapping should only be used for pores etc since the final render will be QUITE large.

2)
Skinning, rigging + blendshapes and all necessary deforming methods to bring the face to life

3)
Deriving an Avatar version from the human model + maybe linking human to avatar skin (by the help of a lot of point/cluster constraints... I have no clue so far)

4)
Texturing and shading with skinshaders + final mapResolution of 8192x8192 pixels to make it a printable version

5)
making use of Maya´s Hair System for all sorts of hair. not only on head but also for the fine hair on skin and feather stuff.

6)
creating fancy HDRI light setups to push it towards photo realism with different environments like the night time jungle and Hallelujah Mountains. Size also around 8192x8192 pixels per facial expression rendering.

Final composition will decorate our living room user added image


everything starts and ends in the right place at the right time.

Last edited by Falott; 16-02-2011 at 10:17 AM.
# 2 16-02-2011 , 09:49 AM
Falott's Avatar
Registered User
Join Date: Jan 2005
Location: vienna
Posts: 1,095

Reference is Key

Well, I´ve freehand modeled heads in the past (with fotoreference ofc) but what I´ve always found to be the hardest thing to overcome is the perspective distortion from a picture taken. you could minimize that perspective distortion by going back as far as possible and zoom in with the camera to the max, but still - getting good resolution pictures with minimal perspective in it (orthographic compact cameras would be great..) isn´t the easiest.

So I was looking for a cheap method to scan the actual real-life geometry. Everyone knows 3D-Scanning can become quite expensive but after some googling I found a small company which scannes your face in 3D and then laseres into a high quality glass cube threedimensionally. Their product sold is this lasered cube and it costs you 20-200 Euros depending of the cube size (2-20 cm). Usually they don´t sell Data and no one asked before I did. So they found it to be interesting. Since it was Valentine´s Day My girl and me got us one of these cubes. The other 10 Scans I let them take from her head were allmost for free. (cost me around 20 bucks). I found this to be the cheapest way to get the most precise reference for starting the modeling process.

Here´s a sample of the lasered glass cube.

Attached Thumbnails

everything starts and ends in the right place at the right time.
# 3 16-02-2011 , 10:06 AM
Falott's Avatar
Registered User
Join Date: Jan 2005
Location: vienna
Posts: 1,095

Reference Scan Data

Here are some results of the scan taken. (I also did let her make some funny faces for later on user added image )

As you can clearely see the disadvantage is, that the scan only works within a range of 180°. And it isn´t that precise in detail since it is after all just a mix of a planar projection of black(shadow) and white(light) lines onto the subject which are then measured in its deviation by some capturing device. you can clearely see the rasterization on the geometry surface. but its just ok for my purpose since I´m most interested in the bigger image concering correct proportioning of the head.

the scans are now used like 3D foto reference. I´m going to block out my raw model on top of that in the beginning. later on foto reference material will be used for detailing as usual.

Attached Thumbnails

everything starts and ends in the right place at the right time.
# 4 16-02-2011 , 03:10 PM
murambi's Avatar
Head of 3d Admedia
Join Date: May 2007
Location: nairobi , ken-ya
Posts: 1,340
wow really looking forward to this

# 5 16-02-2011 , 03:51 PM
Falott's Avatar
Registered User
Join Date: Jan 2005
Location: vienna
Posts: 1,095
thx dude!

first progress on the model for today.
topology/edgeloops will probably change slightly where needed.

I also decided to continue working on the whole face from now, not just the usual way mirroring half a side.

Attached Thumbnails

everything starts and ends in the right place at the right time.
# 6 16-02-2011 , 04:40 PM
honestdom's Avatar
The Nurb Herd
Join Date: Oct 2007
Location: London
Posts: 2,381
i thought you were already further on this?

# 7 19-02-2011 , 08:42 PM
gster123's Avatar
Moderator
Join Date: May 2005
Location: Manchester Uk
Posts: 6,300

i thought you were already further on this?

Eh??

He's retopo scanned data...


"No pressure, no diamonds" Thomas Carlyle
# 8 19-02-2011 , 10:59 PM
honestdom's Avatar
The Nurb Herd
Join Date: Oct 2007
Location: London
Posts: 2,381
there was another thread with a shader problem. i thought it was this model, but finished.

# 9 20-02-2011 , 07:18 AM
Falott's Avatar
Registered User
Join Date: Jan 2005
Location: vienna
Posts: 1,095
ahoi,

yes, I was testing shader settings with an older head, made in the past. this one is built from scratch.


everything starts and ends in the right place at the right time.
Posting Rules Forum Rules
You may not post new threads | You may not post replies | You may not post attachments | You may not edit your posts | BB code is On | Smilies are On | [IMG] code is On | HTML code is Off

Similar Threads