Remember last week, when I said I was working on something really cool for my next post? That was a lie. I had no idea what to do. I still had no idea this evening until about 19:00. Then I slapped something together and shoved it online. But NEXT week, boy, it will make the rapture look like a Tupperware party.

So, I decided to try my hand at a couple of 3D photographs. Snapshots, really. I didn’t use anything to help me align the camera, I just “eyeballed” it and assumed I could fix it up after the fact. The results are not amazing, but a number of the problems could be corrected with another attempt.

These are “cross-eye” stereograms, because that’s my preferred format. I can see that type most easily. Taken from the wikipedia article on the subject:

To view the image cross your eyes until four images appear, then allow the image to converge to a set of three, focusing on the centre image.

They KINDA work. If you concentrate on a particular detail, you can see it, but it’s hard to wander around and explore the image like you can with a good 3D photo, and it is pretty much impossible to take in the whole thing at once. I think the obvious problem is that I used way too much separation between the images. So you can see local details, but there is just not enough similarity between the overall images to see it all at once, or to make looking around easy. They are also not perfectly parallel, not even close. I thought some convergence would help with the almost macro photography style, but I think it works against me.

I took a third set that didn’t suffer from that problem as the subject was much further away. However. the exposure was all wrong. I wanted to silhouette a tree against the setting sun, but I just got a blurry blob. In 3D, the flare from the sun ends up looking like a perfectly flat plane superimposed over the image. Not quite what I was going for but an interesting effect.

Another problem is that I did not make good use of the narrow aspect ratio. I find that narrow stereograms are much easier to see, but because of the huge separation and narrow field of view, there are a lot of background details that are only visible in one or the other image, contributing to the distracting noise when you look at them. Actually, the problem is probably not so much the distance I moved the camera between shots, but rather that I rotated the camera. Next time, I should pick a feature in the distant background to centre the shot on instead of using foreground features.

I think if I attempt this again I should be able to produce better results.

If you wish to use these images, I don’t believe it, and I agree to continue officially disbelieving it even in the face of evidence that you are.