I'm using a modified form of CaptureAndCompressMovie, which provides a
good example of creating a decompression/compression pipeline from the
sequence grabber. Currently the code compresses each frame within the
callback method assigned to the ICMDecompressionSession.
My adaptation has the frames being rendered through a GL context. I'm
experimenting with various filters, which are fairly straightfoward to
chain to the renderering process.
I'd like to apply filters during the encoding process as well. In the
ICMDecompressionSession's callback method, one of the values supplied
is a CVPixelBuffer. I can construct a CIImage from this buffer, set
this image as the "inputImage" value of a filter. I want to transfer
the post-filter image to the compression session. However, the only
way I see how to get the output image from a filter is to draw it to a
context using "CIImage drawImage ...". I suppose the best thing to do
would be to use the CVPixelBuffer as the context, draw the image, then
send the buffer to the encoding method. But I don't see a way to turn
a CVPixelBuffer to a CIContext. Is this the best way for me to be
thinking about this? Can anyone offer any advice?
_______________________________________________
Do not post admin requests to the list. They will be ignored.
QuickTime-API mailing list (email@hidden)
Help/Unsubscribe/Update your Subscription:
http://lists.apple.com/mailman/options/quicktime-api/email@hidden
This email sent to email@hidden
