![](/i/fill.gif) |
![](/i/fill.gif) |
|
![](/i/fill.gif) |
|
![](/i/fill.gif) |
| ![](/i/fill.gif) |
| ![](/i/fill.gif) |
|
![](/i/fill.gif) |
|
![](/i/fill.gif) |
| ![](/i/fill.gif) |
| ![](/i/fill.gif) |
|
![](/i/fill.gif) |
OK, so I have a DVD of some people dancing around. Unfortunately it's
quite difficult to make out who's who. (The camera was struggling with
the light levels.) So I'd like to stick little captions on people so you
can see who they are.
Question: What kind of software would you use for something like that?
I could of course convert the video into a huge stream of still images,
and use something like IrfanView to glue captions onto each frame, one
at a time. However, given several *hours* of film, that's going to take
an absurdly long time, and the captions will jiggle around the screen so
much that they'll be unreadable anyway.
Ideally I'd like to get the captions to float across the screen,
following the people they belong to, but I have absolutely no idea how
to do that.
Any suggestions?
Post a reply to this message
|
![](/i/fill.gif) |
| ![](/i/fill.gif) |
| ![](/i/fill.gif) |
|
![](/i/fill.gif) |
|
![](/i/fill.gif) |
| ![](/i/fill.gif) |
| ![](/i/fill.gif) |
|
![](/i/fill.gif) |
> Ideally I'd like to get the captions to float across the screen, following
> the people they belong to, but I have absolutely no idea how to do that.
>
> Any suggestions?
Without using any fancy auto-tracking software, I'd suggest extracting every
Nth frame, then by hand make a list of names and coordinates for each frame.
Using POV you can read in the list, and using suitable interpolation make an
animation of smoothly floating names (and any other special effects, like
fading in/out). Then use AviSynth or similar to merge the two.
Post a reply to this message
|
![](/i/fill.gif) |
| ![](/i/fill.gif) |
| ![](/i/fill.gif) |
|
![](/i/fill.gif) |
|
![](/i/fill.gif) |
| ![](/i/fill.gif) |
| ![](/i/fill.gif) |
|
![](/i/fill.gif) |
scott wrote:
> then by hand make a list of names and coordinates for
> each frame.
Amazon TURK to the rescue?
--
Darren New, San Diego CA, USA (PST)
Serving Suggestion:
"Don't serve this any more. It's awful."
Post a reply to this message
|
![](/i/fill.gif) |
| ![](/i/fill.gif) |
| ![](/i/fill.gif) |
|
![](/i/fill.gif) |
|
![](/i/fill.gif) |
| ![](/i/fill.gif) |
| ![](/i/fill.gif) |
|
![](/i/fill.gif) |
Invisible wrote:
> Question: What kind of software would you use for something like that?
There are some video capture/creation/playback libraries for Processing:
http://processing.org/reference/libraries/
Some of those can handle face recognition and blob detection. Graphics
in processing are easy to create, so you could track captions to their
owners manually or by using computer vision. Not sure how it would work
out (I would try to break large videos into pieces just in case
something goes wrong), but it's an option.
Post a reply to this message
|
![](/i/fill.gif) |
| ![](/i/fill.gif) |
| ![](/i/fill.gif) |
|
![](/i/fill.gif) |
|
![](/i/fill.gif) |
| ![](/i/fill.gif) |
| ![](/i/fill.gif) |
|
![](/i/fill.gif) |
On 28/10/2010 06:41 PM, Darren New wrote:
> Amazon TURK to the rescue?
LOLrus! :-D
--
http://blog.orphi.me.uk/
http://www.zazzle.com/MathematicalOrchid*
Post a reply to this message
|
![](/i/fill.gif) |
| ![](/i/fill.gif) |
| ![](/i/fill.gif) |
|
![](/i/fill.gif) |
|
![](/i/fill.gif) |
| ![](/i/fill.gif) |