Today has been an up and down day, I had to go into work, so lose 3 hours straight away.

Knowing I was going to be in town I had emailed Jon Maxwell, my contact at the Castle Museum, in the hope he might be about so I can touch base and show him where I am.

Luckily for me he was and I was able to show him the working static model through the Aurasma app and using my trigger image.



I think he was impressed with the 3D model appearing over the trigger, but wanted to know where it would go from there. He is very interested in apps and the Castle doesn’t have one, I told him about my trip to London and the guys there and how they are working it and trying to move maybe towards html rather than proprietary apps, but I know how easy it would be to create an app for them with extra content on their bigger, more popular exhibits, plus a small section for their special collections, it would be simple but a lot of hard work. He would need to look at the viability of implementing it, cost etc, but as I said to him, anything I do for the Bustard he can have as it’s my MA project and my idea to bring to fruition… but as yet no-one in the Norwich Castle Museum is even thinking in this area…

What I realised when I was there and watching him look at it though is the trigger image would need to be much, much bigger if it were to be a floor graphic.


I tried it in situ and also took a photo of two pieces of paper in front of the cabinet, but it still was way off the size I imagined it to be, on reflection it would definitely need to be A2+.

Also, the static model was so dull I really needed to make that animation work.

As soon as I got back home I set about stripping out curves, joints, NURBS, animation etc from the static model which had worked in the Aurasma platform, I thought I had better check it, so popped in a couple of lights and started the export process..

oh no..


Again and again Maya quits on me as I try and export my collada file, I restart my machine, I re-save my scene with a new name, but no luck… Will this ever end…

I look through my scene, surely the  lights and the fact the model is made up of three different materials can’t be a problem,, can it.. I take them all out, it exports…

Back to animating them the simple way, let’s just get some movement in there, I bob the body and dip the head, but without the skeleton and joints, it just looks so awkward… Well, let’s see if we can get some animation through the Aurasma studio… it works, not the best animation, but it works…

Back to Maya, refine the animation slightly, add in some leg movement – this animation will not win any awards, it’s probably the worst animation I have ever done, but I tell you what, it goes through Aurasma again.

It’s looking a little dark, dare I add in a light? I just put in a single spotlight, it goes through…

I try another, the lighting works better now, although the colour is dull, dull, dull.

Looking at the model jigging about in it’s awkward animation I think it’s time to add a little more information into the Aura, it looks good, but an animated bird doesn’t get over what the AR can do, this is more than ‘here’s an animated bird’ job done, it’s about getting over more information, if you need it, if you want it. So I quickly work up a couple of buttons and an information box in illustrator.

GB_info_button GB_info_txtGBOW

I add these into the advanced actionset in the Aurasma studio and check out the result, I like it, but I wish I could get some texture onto the model.

I go back into maya and brave the UV mapper, I re-jig some of my images and fit them onto the shapes, it’s tricky but I get them looking good in Maya.


Upon upload into the Aurasma studio in the 3D preview window it still looks good, I am hopeful.


But when viewed on my iPad the images are all over the place, it’s so disappointing, it would really put the cherry on the cake…

I re-jig and refit and reformat the images, but it doesn’t help on the outcome, I now have 16 slightly different files that I have created in an attempt to get this working.

I look back into the Aurasma guidelines for 3D submission

  • Individual texture maps (.png format) must be of a dimension to the power of 2 (we recommend 128×128, 256×256, 512×512 or 1024×1024). Textures cannot be larger than 1024×1024 pixels.

Ok, I go back into photoshop and make all of my images exactly 512×512 and export as png’s, reattach them to the model and upload, fingers crossed…