Watermarking on key frames

[Lighto]

I would like to inquire on how do I go about using overlay logo onto key frames.

I read about this concept some where in doom9/10 but couldn't find the original thread.

I am quite familiar with using overlay with imagesource but I am not sure on how do I apply and identify the key frames.

:thanks:

[Ghitulescu]

In only 76 threads?

[Lighto]

In only 76 threads?

How exactly did you narrow it down to 76 threads?
Apparently, the keywords that I am capable of providing the forum search function led me to 500 search results in Doom9 alone.

It would be of great help if I can find what I am looking for using the keywords you used to get 76 threads.

[Ghitulescu]

watermarking

[Lighto]

watermarking

Thank you!
Found the thread I was looking for!
http://forum.doom9.org/showthread.php?t=133108&highlight=watermarking

Bad news is that it didn't provide me with the method to identify key frames using avisynth, I would like to know if it's possible to do so using avisynth as well.

[Sir BlunderBrain]

As far as I know AviSynth doesn't really have any knowledge of "key frames", it only deals with raw uncompressed frames.

If you want to process the frames from your source that were originally key frames you might be able to do something with ConditionalReader (http://avisynth.org/mediawiki/ConditionalReader) provided that you can find some tool to extract the key frame numbers...

If you want to process the frames that will become key frames that's something else. That decission is made by your target encoder and not by AviSynth so there is no way for your script to know what the encoder will do.

If your encoder has some option like "force key frame every xx frames" you might be able to predict at least some of the key frames, but it may also have a negative impact on the filesize.

[Lighto]

As far as I know AviSynth doesn't really have any knowledge of "key frames", it only deals with raw uncompressed frames.

If you want to process the frames from your source that were originally key frames you might be able to do something with ConditionalReader (http://avisynth.org/mediawiki/ConditionalReader) provided that you can find some tool to extract the key frame numbers...

If you want to process the frames that will become key frames that's something else. That decission is made by your target encoder and not by AviSynth so there is no way for your script to know what the encoder will do.

If your encoder has some option like "force key frame every xx frames" you might be able to predict at least some of the key frames, but it may also have a negative impact on the filesize.

Instead of using the encoder to "force key frame every xx frames", is it better off using stickboy's ApplyEvery plugin?

[Sir BlunderBrain]

Unless you have some very specific reason to only watermark key frames I would figure out some other algorithm to determine where to put it, like a simple "every x" frames (ApplyEvery) or some random distribution. It would probably save you alot of headache.