On 22/01/13 17:17, James Morrissey wrote:
Hi,
I am trying to put together a presentation for a conference. The
conference is in Australia and rather than flying there i am trying to
make a video of my presentation, following which i will skype into the
QA plenary.
I had thought of simply reading the paper, but worry this might get a
little boring for the audience. So now i am thinking of including some
impress slides into the presentation. With that in mind i also don't
want to simply have my voice running over an impress slide.
So i have thought of trying to make a screencast with an image from my
webcam running in one of the corners. This seems easy enough, just set
cheese (or some equivalent) running and record my desktop. The problem
though is that i'd ideally like to be able to run impress in full
screen mode. Doing so, however, covers up cheese - even if i set it to
'always on top'.
In a perfect world i'd also be able to resize the video of myself,
during the video, so that when the slide is more relevant to what i am
saying it dominates the picture and when what i am saying is most
relevant, my face does.
The solution i thought of was to run cheese on one workspace and
recordmydesktop on another, where Impress is running in full screen
mode. I'd then give the talk, clicking through impress, with both
recordings running. Following that i'd open one instance of Totem and
another of VLC, one showing the cheese video of me and the other the
screencast of Impress. I'd then record that desktop, resizing the
cheese video, of myself, at the different parts of the talk.
Unfortunately it seems that running recordmydestop means that cheese
can't record from my webcam - it can show the image, but crashed when
i hit record and now can't find my webcam. The same thing has happened
with Kamoso which also now can't find the webcam.
I was thus thinking that i could try and use a cam corder, and take a
video of me giving the talk before taking a screen cast of the impress
presentation, using the audio from the talk as my cue for clicking
through the slides. I'd then stick the two of them together, as
described above, recording the two videos playing on top of one another.
Rather than do that, what about recording yourself with Cheese first and
save the video to a file? No need for a camcorder. Then re-run the audio
and click through the slides. Then save the screencast of the slides to
another video file.
Finally combine both videos using gstreamer picture-in-picture
compositing abilities, as explained here:
http://www.oz9aec.net/index.php/gstreamer/347-more-gstreamer-tips-picture-in-picture-compositing
You could probably use a video editor to do this as well.
Cheers,
Bruno
--
ubuntu-uk@lists.ubuntu.com
https://lists.ubuntu.com/mailman/listinfo/ubuntu-uk
https://wiki.ubuntu.com/UKTeam/
