I heard about these recently and it got my interest. I wanted to create them as volumes in Houdini. I chose the Korean Visible Human because their slices were 0.1 mm in thickness as opposed to 0.33 mm used by US for their Visible Human project . Also their data set isn't open source.

You can find the KVH dataset here:
https://ia600406.us.archive.org/31/items/VisibleKoreanHumanImages/ [ia600406.us.archive.org]

They only share every 5th slice though, which is sufficient as I just wanted to see how it looks in Houdini.

Anyway here is how it looks in the viewport:



I didn't use all slices because cooking for 700 slices (out of 1700) took 30 mins.

I didn't spend much time setting up the volumes but I probably spent an hour keying and masking in Combustion.

If there are more optimal/faster ways to create a volume from a set of sliced images, please let me know

Wow that's strange

Thanks I know it's a little creepy

Oh wow, thanks for posting this! I had not heard of it before.

I wanna try it too, someday soon, when I'm done with my other projects

Np, for some interesting info you can check out this wiki:
http://en.wikipedia.org/wiki/Visible_Human_Project [en.wikipedia.org]

There is also a PDF released by the Koreans who did the KVH project and it actually describes the whole process in detail, i.e. how they preserved the guy, how the “slicing” was performed, how it was photographed, etc.

For the curious minds, in case you are wondering if they used lasers, or extremely sharp blades, hacksaw, or a band saw, they didn't. They used a milling machine 8)

I find it a little strange that they even mentioned that the whole process was completely destructive to the specimen

Looking fresh! Have you thought about porting this over to a shader or volume procedural? This would defer your cook times to the render and you wouldn't need to write or load a volume.

No I haven't but is there a simple example I can use as a template? I am not sure how to go about doing it as a volume procedural.

In my setup, I reckon most of the time is spend by reading the image files for each voxel. Only way I found to set all voxels of a particular “slice” at once is by writing a python SOP using setAllVoxelsFromString.

I didn't give that a try but that should provide a significant speed up.

Hope this example will help you.

Thanks James, I will check it out when I am back from vacation

I did something similar RE reading volume slices in, but I output it to minecraft :3

yeeeeeaaaaaaaaaah probably should have resized those

pusat
Np, for some interesting info you can check out this wiki:
http://en.wikipedia.org/wiki/Visible_Human_Project [en.wikipedia.org]

There is also a PDF released by the Koreans who did the KVH project and it actually describes the whole process in detail, i.e. how they preserved the guy, how the “slicing” was performed, how it was photographed, etc.

For the curious minds, in case you are wondering if they used lasers, or extremely sharp blades, hacksaw, or a band saw, they didn't. They used a milling machine 8)

I find it a little strange that they even mentioned that the whole process was completely destructive to the specimen

Very Useful reference ,Thanks

Thanks That minecraft thing also looks pretty cool.