FredThompson
14th October 2003, 19:42
This title stinks, nothing I can think of adequately describes what I'm trying to do in just a few short words.
I've tried a number of ways to convert PowerPoint presentations to DVD. S-Video out to a DV camcorder is ok but there are some problems. What I'd really prefer to do is simulate the presentations with AviSynth. This will allow far more quality and control.
With my presentations, each "slide" is based on a template with a consistent static element, some variable text, either a still image or a video segment, and a 5 second delay which starts immediately for static images or after a video segment runs. I know how to simulate most of this with AviSynth EXCEPT the timing of variable-length video segments.
I'm looking for a way to construct frames such that the frame numbers which correspond to the begining of each "slide" are automatically calculated.
As an example pseudo-AviSynth code:
frame 0
...procedure to show static elements
...code to show "slide" specific static elements
...code to show still image or video segment
frame ?
...etc
frame ?
...etc
If the slide is to show a static image, the offset to the following frame should be easily calculated. How would this be done with an embedded video segment?
Ideally, the code for specific "slides" could be modified without having to hardcode frame ranges.
Embedded video should be available for processing through a script for clean up (interlaced source->kernaldeint->noise filters, etc.)
Perhaps one solution would be to use a label for the "start" of each "slide".
Has anyone done something like this? Would someone have suggestions how to code such variable-length frame ranges with interdependencies based on embedded segment length?
I've tried a number of ways to convert PowerPoint presentations to DVD. S-Video out to a DV camcorder is ok but there are some problems. What I'd really prefer to do is simulate the presentations with AviSynth. This will allow far more quality and control.
With my presentations, each "slide" is based on a template with a consistent static element, some variable text, either a still image or a video segment, and a 5 second delay which starts immediately for static images or after a video segment runs. I know how to simulate most of this with AviSynth EXCEPT the timing of variable-length video segments.
I'm looking for a way to construct frames such that the frame numbers which correspond to the begining of each "slide" are automatically calculated.
As an example pseudo-AviSynth code:
frame 0
...procedure to show static elements
...code to show "slide" specific static elements
...code to show still image or video segment
frame ?
...etc
frame ?
...etc
If the slide is to show a static image, the offset to the following frame should be easily calculated. How would this be done with an embedded video segment?
Ideally, the code for specific "slides" could be modified without having to hardcode frame ranges.
Embedded video should be available for processing through a script for clean up (interlaced source->kernaldeint->noise filters, etc.)
Perhaps one solution would be to use a label for the "start" of each "slide".
Has anyone done something like this? Would someone have suggestions how to code such variable-length frame ranges with interdependencies based on embedded segment length?