x404.co.uk
http://x404.co.uk/forum/

Anaglyph Experiment - which is better to view?
http://x404.co.uk/forum/viewtopic.php?f=17&t=5419		 Page 1 of 1
Author:  paulzolo [ Sun Jan 10, 2010 7:35 pm ]
Post subject:  Anaglyph Experiment - which is better to view?
I’ll post this here to avoid confusion with the other thread. I have a couple of techniques available to me for building anaglyphs. I have my own private idea of which is better, but I want more input on this. You will need red/cyan 3D glasses.

Image A:
Image
(Straight from the camera - both left & right images overlaid directly)

Image B:
Image
(Post work done to adjust the position of images when overlaying)

Both images produce a 3D effect.

My question - which is better? Not only for 3D but also for ease of viewing? Should you have to “work” to view such images or should they just “pop”?
Author:  belchingmatt [ Sun Jan 10, 2010 7:50 pm ]
Post subject:  Re: Anaglyph Experiment - which is better to view?
Where can I pick up some decent glasses, preferably free?
Author:  paulzolo [ Sun Jan 10, 2010 9:20 pm ]
Post subject:  Re: Anaglyph Experiment - which is better to view?
belchingmatt wrote:
Where can I pick up some decent glasses, preferably free?


Not seen any on magazines recently. There are sites out there that sell them for a couple of quid upwards.

You could make your own: http://www.haworth-village.org.uk/3d/3d-glasses.asp
Author:  Fogmeister [ Sun Jan 10, 2010 9:27 pm ]
Post subject:  Re: Anaglyph Experiment - which is better to view?
For me image B stands out a lot better with both the 3D effect AND the colours.

Image A looks a lot more black and white and a lot flatter.
Author:  nickminers [ Sun Jan 10, 2010 9:52 pm ]
Post subject:  Re: Anaglyph Experiment - which is better to view?
B works far better for me too.
Author:  rustybucket [ Sun Jan 10, 2010 11:21 pm ]
Post subject:  Re: Anaglyph Experiment - which is better to view?
In image A I'm finding a hell of a lot of detail leaking between eyes. I think it's because, (although there's only a little red leaking into the right eye and no blue into the left), the luminosity is changed when the colours are separated on a dark or light bagkground. Consequently I still get a a "sort-of" 3D thing going on but I'm having to work a whole lot harder to resolve the image.

In image B the effect does work a whole lot better. However I'm still getting some leakage mainly on the left (red) channel. This most apparent on the trims on the car in front and the Bairstow Eves branch in the back.
Author:  paulzolo [ Mon Jan 11, 2010 12:09 pm ]
Post subject:  Re: Anaglyph Experiment - which is better to view?
rustybucket wrote:
In image A I'm finding a hell of a lot of detail leaking between eyes. I think it's because, (although there's only a little red leaking into the right eye and no blue into the left), the luminosity is changed when the colours are separated on a dark or light bagkground. Consequently I still get a a "sort-of" 3D thing going on but I'm having to work a whole lot harder to resolve the image.

In image B the effect does work a whole lot better. However I'm still getting some leakage mainly on the left (red) channel. This most apparent on the trims on the car in front and the Bairstow Eves branch in the back.


Some of this may be down to the filters you are using. I have two pairs of red/cyan glasses, and one does the job far better than the other. The worse pair came with a book of optical illusions. The other I got at the cinema when Red Bull were doing 3D ads for their AirRaces a few years back. The RedBull pair are the better of the two, but there is still “ghosting” at times - which is unavoidable. The BairstowEves banner is red - any reds or blues can cause odd behaviour - it’s usually best to avoid them, and if I were doing more than a quick job on these, i’d make the red less so before making the anaglyph.

Image A is the two pictures from the camera plonked down on top of each other with no further adjustment. Obviously, the offset of the two images is determined by the distance the lenses are from each other. I have not ability to move them closer. On the 3D screen on the back of the camera, the image works well, and you would expect the camera to be tuned for this. Fuji also do a 3D display and a 3D printing service - and I would expect those to be adjusted for the camera too.

Image B has had extra adjustments done by me in an application that strives to restore colours and give control over image placement. By reducing the overlap, you lose a lot of the ghosting, get better colour reproduction and a for more comfortable experience. I find this to be a far better image than the first for these reasons.

For those who are interested, the app I used to create Image B is called 3D Slide Maker and is available for the Mac from: http://www.thebox.myzen.co.uk/Steroscop ... tware.html

It’s a spot buggy in places, but on the whole for a free app, it’s pretty damn good.

EDIT: Just realised - the writer of that software used to write for Acorn User - Mike Cook. He used to do a lot of interesting hardware stuff.
Page 1 of 1 All times are UTC
Powered by phpBB® Forum Software © phpBB Group
https://www.phpbb.com/