Hat and bag render tests

Using the 3d paint tool I quickly drew up some childish bug designs on the bag using Phils bag drawings as a reference. I had a little trouble with this at first because it was mirroring what I was painting despite the reflection setting being off. Alex Hulse then helpfully pointed out that it's because the UV map was overlapping, therefore so would the drawing. Here it is anyway, and as promised I did a quick compositing test with the renders as well to enhance the quality.

Right, now that's sorted out I believe it's blendshape time, and Phil has posted expression sheets so I have a good reference!

UV mapping on the hat

With help from Jaimee Perrin, Phil and myself set out to work out and do the uv mapping for the ladybug pattern on the hat. It eventually worked out ok, although we did have the odd problem with getting it right so the spots didn't deform tragically. As soon as I have acess to after effects I want to post the images after it's been treated with the same effects that the last post proved to work.

This now means that I can spend the weekend sorting out blendshapes and the facial control pannel so I can finally get around to skinning. With any luck Phil should be animating by next week and I can turn my attention fully to Pocahontis Billiams.

We also had the formative assesment today, which proved to be a very positive one. I think we are on the right track here. The only real negative commentery was to do with my lack of personal creative interpretation. While we did come to an understanding that for the specialist areas I have undertaken there is not a lot of room to personal creativitiy, (Phil designs and I make it happen in Maya), I should perhaps still make an extra effort to deal with this!

Colouring/rendering experiments!

Based on Pocoyo we decided that texturing was only going to over complicate a design that benefits from simplicity, and have therefore decided to turn our attention to rendering and After Effects to give it sparkle! Obviously we are still going to need to delve into UV mapping to texture the hat with the ladybug design, and will need to learn how to paint the bug designs onto the bag, but other than that anything else is unnecessary.

Above is a quick test: the top image is the basic colour layer rendered out in mental ray, with the colours unadjusted. The below image is the colour layer with an occlusion layer, overlayed in after effects. The image was further altered with the colour balance, and brightness/contrast tools. As you can see the difference is immense, and it goes to show what more we could accomplish if we used more than the very basics of the software.

Importing the head

Just a quick update, I cheated a little here and imported the rough head model from one of Willies early files. Figured it would save some time and enable me to focus my energies on the difficult part, which will be the hair. It could still use a little reshaping perhaps.

Pocahontis Billiams!!

Basically the body for Willies little sister, Pocahontis is modelled to the point of completion. (Except from any revisions that may need to be made upon Phils review.) This is a far simpler design so has caused nowhere near as many issues as Willie did. Having said that I havn't attempted the head yet, which I'm predicting will be difficult- mainly because of the girly hair.

Antennae problem solved

The problem was very easy to solve. It was a simple case of parenting the curves from the IK spline handles to the head controller. I then parented the antennae controllers to the head controller as well, which means when rotating the head they automatically do the secondary animation- you can see from the clip how it works in a kind of whip like motion. Of course this might not be ideal for posing, but you can still control them manually so it's no problem.

Also, Phil, you wanted to know if they are flexible enough to touch the cap. Straight answer is yes they are.

nearly rigged!

I got the driven keys set for the fingers, and it's all working fine. Had a bit of trouble with the right thumb, but it was only a simple case of re-orientating the joints before it was working fine. I also included a finger spread attribute. I made sure the joints were large enough to be manipulated manually through the geometery, as set driven keys disables the ability to set up attributes for the individual rotations. Of course with four fingers and a thumb, I wouldn't want to overload the channel box with attributes.

As for the antennae, I set them up with IK splines and one locator at the top to control the curve. I went with one controller as it gives it a more springy effect, rather than several controllers which would make it over-flexible. As you can see from the video however, it means that the antennae stay in a fixed position when they should be following the movements of the head. I need to resolve this, and quickly as we are already way behind schedule.