• Open Menu Close Menu
  • Apple
  • Shopping Bag
  • Apple
  • Mac
  • iPad
  • iPhone
  • Watch
  • TV
  • Music
  • Support
  • Search apple.com
  • Shopping Bag

Lists

Open Menu Close Menu
  • Terms and Conditions
  • Lists hosted on this site
  • Email the Postmaster
  • Tips for posting to public mailing lists
Re: CoreVideo to pixel level data (speed path problems)
[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: CoreVideo to pixel level data (speed path problems)


  • Subject: Re: CoreVideo to pixel level data (speed path problems)
  • From: "douglas a. welton" <email@hidden>
  • Date: Mon, 16 Oct 2006 21:11:53 -0400

Jeremy,

(Hopefully I have read your post and understood it correctly...)

Check out the FrameReader object in the QCTV sample code. If I remember correctly, if manipulates the the pixels from the CV buffer directly into an image description... (No CIImage, No NSImage, No bitmap reps)

Have you checked out the CIVideoDemoGL sample code? The - exportMovie: routine in this sample uses the MovieExporter API and grabs CV buffer data directly into an image description as well. If you just want the Red pixels from a buffer, you should be able to stick the CIColorCube filter into this code and get (something close to) what you want.

Hope that helps.

later,

douglas

On Oct 16, 2006, at 8:03 PM, Jeremy Faller wrote:


I'm making a laser range finder: (http://www.pages.drexel.edu/ ~twd25/webcam_laser_ranger.html)


I have written the SLANG filter for red filtering, and it works fine. However, I think QuartzComposer SLANG filters fail for this type of application.


On Oct 16, 2006, at 8:01 PM, John C. Randolph wrote:

I tend to do pixel-bashing with the help of Quartz Composer these days. It's pretty easy to extract whatever pixel-level info you want from a video stream using a QC "Core Image Kernel" patch with a bit of GL SLANG code.

What are you trying to do with this red data?

-jcr

On Oct 16, 2006, at 4:39 PM, Jeremy Faller wrote:

All:


I am trying to take iSight data, and get the bitmap data for post-processing. I am currently using the WhackedTV example from Apple as a base, and it converts CVPixelBufferRef-->CIImage(s)-- >NSImage-->NSBitmapImageRep, but these operations are expensive and slow. (I am seeing that converting the images to NSBitmapImageRep can peg a processor at low frame rates. All of the time is spent in [NSImage drawImage:], the final step to convert NSImage(s) to NSBitmapImageRep.)


I am interested in knowing if there are any quick ways to access the pixel level data in any of the data structures I am using? Does anyone have any example code?

As a (possibly useful) aside, I am only interested in the red data (from RGB) of the pixel data. If it helps, I would be glad to sacrifice G/B/Alpha for just the red.


Thanks in advance.



John C. Randolph,
VP, Engineering
Stealth Imaging, Inc. <email@hidden>





_______________________________________________
Do not post admin requests to the list. They will be ignored.
Cocoa-dev mailing list (email@hidden)
Help/Unsubscribe/Update your Subscription:
40earthlink.net


This email sent to email@hidden

_______________________________________________ Do not post admin requests to the list. They will be ignored. Cocoa-dev mailing list (email@hidden) Help/Unsubscribe/Update your Subscription: This email sent to email@hidden
References: 
 >CoreVideo to pixel level data (speed path problems) (From: Jeremy Faller <email@hidden>)
 >Re: CoreVideo to pixel level data (speed path problems) (From: "John C. Randolph" <email@hidden>)
 >Re: CoreVideo to pixel level data (speed path problems) (From: Jeremy Faller <email@hidden>)

  • Prev by Date: Re: Two TreeControllers, one ArrayController and the ContentSet binding...
  • Next by Date: Re: playing a quicktime movie in a QTMovieView
  • Previous by thread: Re: CoreVideo to pixel level data (speed path problems)
  • Next by thread: PDF Kit [PDFPage setBounds:forBox:] not working!
  • Index(es):
    • Date
    • Thread