hello. I want to get the pixels of a video capture from vvvv and send it to processing as rgb strings via udp. udp communication with processing works great. My only problem is to get the pixels from video. I thought I should do: videoIn -> videoTexture -> quad. now that I can think of my video feed as a normal texture, is there a way to extract the pixel or rgb information from a regular texture?
or directly send the output of AsString (EX9.Texture) to processing and parse it there. sending a 320240 video encoded as rgb ascii would be 32024033 = 691200 characters per frame. anyway, depending on what you like to do with the pixels, writing a pixelshader and keep all in vvvv is at least 10 times faster than sending the pixels to processing an do something there on the CPU.
thank u all. i guess you are right tonfilm in that it would be really stupid to send data to processing. well then my journey to shaders begins…wish me luck:)
nevertheless, I tried joreg’s suggestion, but I think pipet is broken. It doesn’t work with filetexture and videotexture.
try beta12. pipet works there. in beta13 pipet seem broken with videotexture only.
in beta13 pipet seem broken with videotexture only
??? for me it works.
ai, what i meant to say was: broken with filetexture only
hi again. I succeded in getting pixels from videotexture with pipet. However, although i change the x in pipet, it doesn’t scan all the pixels horizontally. It gets stuck with the lower right pixel. On the contrary, when I use filetexture with pipet (beta12), and change x, it does scan all the pixels horizontally.
here is my patch. any help would be great.
blober01.v4p (8.0 kB)
connect videoin to videoout for the most basic test to see video. check both splitcam-options on videoin with this setup.