Hacked some python script together for parsing this and generated some HTML file with the data out of it.
Still not sure when to change the frame, the whole thing is of by some pixels.
Should i post it, not sure about the legal component. I'm from germany so reverse engineering is allowed in some cases, not sure if this is covered here.
Surely those are just standard depth-maps as used every day in VFX for compositing? Each pixel has a z-depth as to how far it is from the camera, and using standard compositing software (Nuke) you can blur just a slice (depth-wise) of the image based on the z ranges?
Comments
http://lytro.com/gallery/fieldsofgold2.dat http://lytro.com/gallery/lytro_50_00084.dat http://lytro.com/gallery/lytro_50_00087.dat http://lytro.com/gallery/lytro_50_00090.dat
custom binary format?
And I suppose this is the key
http://lytro.com/include/refocus.swf
Hacked some python script together for parsing this and generated some HTML file with the data out of it. Still not sure when to change the frame, the whole thing is of by some pixels.
Should i post it, not sure about the legal component. I'm from germany so reverse engineering is allowed in some cases, not sure if this is covered here.
So here is the parser script. The example.html is very messy sorry for that. https://gist.github.com/997861
This is HN! Post the darn thing!!
Surely those are just standard depth-maps as used every day in VFX for compositing? Each pixel has a z-depth as to how far it is from the camera, and using standard compositing software (Nuke) you can blur just a slice (depth-wise) of the image based on the z ranges?
Presumably the data format is based on their light field viewer: http://lightfield.stanford.edu/aperture.swf?lightfield=data/...
(note this downloads 12 Mb of data)
You can look at the source: http://lightfield.stanford.edu/aperture.html