I need to take the output of pix_video and reduce the resolution considerably, to something like 16 x 12. Can anyone suggest a reasonable way to do that?
While on the subject, does anyone have good pointers to doing something in the general ballpark of track-them-colors (Director) or Cyclops (Max)?
Many thanks.
- martin
This email has been independently scanned for viruses and any virus software has been removed using McAfee anti-virus software
Martin Dupras wrote:
I need to take the output of pix_video and reduce the resolution considerably, to something like 16 x 12. Can anyone suggest a reasonable way to do that?
this has been on my todo-list for ages ;-(
While on the subject, does anyone have good pointers to doing something in the general ballpark of track-them-colors (Director) or Cyclops (Max)?
i have no idea what they do, but from the director-name, i guess it has something to do with color.
[pix_blob] (0.90), [pix_multiblob] (CVS) do blob-tracking based on the color-planes (well, this is true for [pix_blob]; [pix_multiblob] only considers the luminance) so you'll have to transform the color-to-track-information to be directly accessible via linear combination ([pix_blob]) of the color-planes (or via the luminance-plane with [pix_multiblob]) i use [pix_rgb2hsv] for this, but this surely eats a considerable amount of your cpu...
mfg.asd. IOhannes
IOhannes m zmoelnig wrote:
Martin Dupras wrote:
I need to take the output of pix_video and reduce the resolution considerably, to something like 16 x 12. Can anyone suggest a reasonable way to do that?
this has been on my todo-list for ages ;-(
oh, and i forgot to mention: you should be already able to do this with [pix_resize]; if you don't give any arguments, it will enlarge an image to have a 2^n-size. (e.g. a 200x100 image will be resized to 256x128) however, you can give arguments to define another size, e.g.: [pix_resize 16 12] will resize the image to guess what... note however that this uses gluScaleImage() which might bot be the fastest algorithm in the world.... (probably not a problem when downsampling)
mfg.asd.r IOhannes
That's *really* good to know, and it wasn't in the helpfile last time I was there... hint, hint.
-Ian
IOhannes m zmoelnig wrote:
IOhannes m zmoelnig wrote:
Martin Dupras wrote:
I need to take the output of pix_video and reduce the resolution considerably, to something like 16 x 12. Can anyone suggest a reasonable way to do that?
this has been on my todo-list for ages ;-(
oh, and i forgot to mention: you should be already able to do this with [pix_resize]; if you don't give any arguments, it will enlarge an image to have a 2^n-size. (e.g. a 200x100 image will be resized to 256x128) however, you can give arguments to define another size, e.g.: [pix_resize 16 12] will resize the image to guess what... note however that this uses gluScaleImage() which might bot be the fastest algorithm in the world.... (probably not a problem when downsampling)
mfg.asd.r IOhannes
PD-list@iem.at mailing list UNSUBSCRIBE and account-management -> http://lists.puredata.info/listinfo/pd-list
-
Ian Smith-Heisters -
IOhannes m zmoelnig -
Martin Dupras