T1 and T2 Endoskeleton Research Thread

[ghostman]

ghostman: Looking good. The jaw looks positioned a little crooked. Too much towards the right (left in the pictures.) I think if you move it a little to the other side, so the bottom teeth center line up with the top teeth center more, it would look better.

Since it's been so diffuclt to find castings or scans of the complete SWS spine, we may get it sculpted by a talented artist.
ghostman, I was wondering if we would be allowed to use your spine photogrammetry model as visual and 3D reference, as even though it isn't 100% it still has a cool look and some accurate shapes here and there that can assist us in getting this made properly? If you get the Cinemaquette spines recreated with photogrammetry, would we be allowed to use those too, as reference?

Nokitten: That's a really impressive looking piece. Great work.

I think I did a good job of photogrammetry.
But upper jaw is a little rough compared to the lower jaw.
I took this photo at ISO 100 and f/11.
I'll stop down to f/22 to increase the depth of field and try again.
I'm worried about blurring because the shutter speed slows down as I stop down the aperture...
I will try again on a brighter cloudy day when the light is more diffused.

Of course, I plan to shoot Cinemaquette after this.
But getting this skull in print-ready condition first.

 

[DeltaDesignRus]

I think I did a good job of photogrammetry.
But upper jaw is a little rough compared to the lower jaw.
I took this photo at ISO 100 and f/11.
I'll stop down to f/22 to increase the depth of field and try again.
I'm worried about blurring because the shutter speed slows down as I stop down the aperture...
I will try again on a brighter cloudy day when the light is more diffused.

Of course, I plan to shoot Cinemaquette after this.
But getting this skull in print-ready condition first.

Before you process the scans obtained from Metashape, you should try RealityCapture; maybe it will provide better results, and then less manual work will be required. At the same time, you will demonstrate how both programs handle the same dataset.

 

[Too Much Garlic]

Spent 4 hours fixing the toes into single solid pieces... made the mistake of not saving along the way... and suffered for it. The program stalled and all that work was lost. Got really demoralized, so haven't done anything since. It was my own fault. It still sucks.

 

[ghostman]

The skull is on loan and must be returned after the work is completed first.
Cinemaquette will be filmed afterwards.
Last time I shot 24 pictures per round, this time I plan to shoot 48.
The nostrils will be better than last time.
The sputtering paint was thick in some areas, so I reapplied some of it.
The contrast was high because it was taken on a sunny day.
I will take more pictures on an overcast day when the light is more diffused.

I tried photogrammetry only on the lower jaw with RealityCapture.
The calculation is indeed very fast.

 

[DeltaDesignRus]

I've reached the brass casting. Since my skull is at a reduced scale (approximately 85%), I had to make these parts at the same scale as the skull. Cast in brass, after processing, they will be nickel-plated. Not all parts are here yet; the rest will be cast another time. This was just a trial, but it exceeded my expectations; the quality is quite good.