## jmol-users

 [Jmol-users] Re: Stereo Images in Jmol (Miguel) From: - 2005-07-28 21:05:41 ```Hi with regards to the red/green stereo images i think you simply put the left eye is rendered transparent to green, and the right eye image transparent to red this way the colored lense filter out the apropiate image. The disadvantage is you cant view colors, but the big plus is that the glasses a very cheep (50p each) so users can benifit from real 3d very easily. Any thanks for even considering it. Andrew Milsted ```
 [Jmol-users] Re: Stereo Images in Jmol (Miguel) From: Miguel - 2005-07-29 00:19:33 ```> =5Bgetting list mail but still not able to reply on-list=5D > > On =5B2005-Jul-28=5D Miguel wrote: >> I am not exactly sure what is going on ... but here is my theory ... >> >> One channel is red >> >> The other channel is cyan ... green + blue ... I observe that the the >> byte values for G and B are always exactly the same. >> >> I guess that they are building the left and right images independently= >> as >> grey scale, then extracting the R from one image and the G+B from the >> other. > > I think =22imposing=22 rather than =22extracting=22. When I get back in= the office > I can send an image with almost a complete separation between two spher= es > and showing the colour content. Where you sit depends upon where you stand :-) For each pixel I will have two RGB values, left and right. I need to combine them to form one RGB value. >> I need to think about how to implement this ... > > I think you just need to calculate the two images > (either via rotation or by the asymmetric frustrum) I am still using rotation ... have not implemented asymmetric frustrum. > and then impose the colour value + intensity value (lightness) > corresponding to the base greyscale pixel in that image. Each 'greyscale pixel' will be an RGB value. But, since they are 'greyscale' the component bytes R, G and B will actually be equal. This means that we can think of left and right as byte= values in the range 0-255, not RGB values. The final result is composed of the R byte from the left and the G+B byte= s from the right (or vice versa) If you want red/blue instead of red/cyan then leave the green component set to 0. Miguel ----- Open Source Molecular Visualization http://www.jmol.org miguel=40jmol.org ----- ```
 Re: [Jmol-users] Re: Stereo Images in Jmol (Miguel) From: Miguel - 2005-07-28 22:22:45 ```> with regards to the red/green stereo images i think you simply put the left eye is rendered transparent to green, and the right eye image transparent to red this way the colored lense filter out the apropiate image. Conceptually, that is correct. But the definition for how to *implement* it needs to be taken to a lower= level. This is the last exchange I had with someone else off-list: > I've attached two tiff files. The first (mono) is how the molecule > looks in the normal display and the second image shows how it appears > when the red-blue anaglyph is formed. I am not exactly sure what is going on ... but here is my theory ... One channel is red The other channel is cyan ... green + blue ... I observe that the the byt= e values for G and B are always exactly the same. I guess that they are building the left and right images independently as= grey scale, then extracting the R from one image and the G+B from the other. I need to think about how to implement this ... Miguel ```

## Get latest updates about Open Source Projects, Conferences and News.

Sign up for the SourceForge newsletter:

No, thanks