WEBVTT

00:00:00.000 --> 00:00:05.340 align:middle line:90%


00:00:05.340 --> 00:00:07.590 align:middle line:84%
We have been looking
at how we can

00:00:07.590 --> 00:00:11.160 align:middle line:84%
do motion capture of musicians
and also of perceivers.

00:00:11.160 --> 00:00:15.690 align:middle line:84%
But some people are actually
doing it even more complicated

00:00:15.690 --> 00:00:18.030 align:middle line:84%
than that, and trying
to combine doing

00:00:18.030 --> 00:00:21.870 align:middle line:84%
motion capture of many
people at the same time.

00:00:21.870 --> 00:00:24.450 align:middle line:90%
And Mari, you're one of them.

00:00:24.450 --> 00:00:27.620 align:middle line:84%
Tell me a little bit
about your studies.

00:00:27.620 --> 00:00:28.120 align:middle line:90%
OK.

00:00:28.120 --> 00:00:34.380 align:middle line:84%
Yeah, so I have done recordings
of both musicians and dancers

00:00:34.380 --> 00:00:37.770 align:middle line:90%
in folk music genres.

00:00:37.770 --> 00:00:39.900 align:middle line:90%
I recorded them simultaneously.

00:00:39.900 --> 00:00:40.830 align:middle line:90%
Wow.

00:00:40.830 --> 00:00:43.110 align:middle line:84%
But that must be quite
complex, because you have then

00:00:43.110 --> 00:00:46.230 align:middle line:84%
people moving around,
and sitting still.

00:00:46.230 --> 00:00:50.340 align:middle line:90%
Well, how did it go?

00:00:50.340 --> 00:00:57.510 align:middle line:84%
Yeah, so we did a lot
of planning, and to plan

00:00:57.510 --> 00:01:00.180 align:middle line:84%
how to place the cameras
and stuff like that,

00:01:00.180 --> 00:01:04.500 align:middle line:84%
because as you said, the dancers
are really moving around.

00:01:04.500 --> 00:01:09.240 align:middle line:84%
And in particular, one of the
styles that I've been recording

00:01:09.240 --> 00:01:15.360 align:middle line:84%
is the Norwegian folk music
style called Telespringar.

00:01:15.360 --> 00:01:17.970 align:middle line:84%
There you have it looking
at couples dancing together.

00:01:17.970 --> 00:01:19.440 align:middle line:90%
That must be very difficult.

00:01:19.440 --> 00:01:20.640 align:middle line:90%
Yeah.

00:01:20.640 --> 00:01:26.070 align:middle line:84%
And also, it's a challenge
because they dance

00:01:26.070 --> 00:01:28.000 align:middle line:90%
very close to one another.

00:01:28.000 --> 00:01:30.360 align:middle line:84%
So, you really have
to really think

00:01:30.360 --> 00:01:36.630 align:middle line:84%
about where to place the markers
to prevent occlusion, but also

00:01:36.630 --> 00:01:39.060 align:middle line:90%
how to place the cameras.

00:01:39.060 --> 00:01:45.150 align:middle line:84%
But then the musicians were
still sitting quite still.

00:01:45.150 --> 00:01:48.030 align:middle line:84%
But you managed then to do
them both, one musician sitting

00:01:48.030 --> 00:01:50.820 align:middle line:84%
still, and dancers moving
together, and spinning around.

00:01:50.820 --> 00:01:52.770 align:middle line:84%
And also in a
fairly large space.

00:01:52.770 --> 00:01:55.300 align:middle line:84%
How big of spaces
were you working in?

00:01:55.300 --> 00:01:56.670 align:middle line:90%
Yeah.

00:01:56.670 --> 00:01:58.620 align:middle line:90%
It wasn't that big, actually.

00:01:58.620 --> 00:02:04.110 align:middle line:84%
But it was big enough for
the dancers to move about.

00:02:04.110 --> 00:02:07.301 align:middle line:84%
And then in terms of
putting on the markers,

00:02:07.301 --> 00:02:09.509 align:middle line:84%
I guess if you have markers
on the front for example,

00:02:09.509 --> 00:02:10.570 align:middle line:90%
they would be covered up.

00:02:10.570 --> 00:02:12.000 align:middle line:90%
Or how did you handle that?

00:02:12.000 --> 00:02:14.640 align:middle line:84%
No, that actually
worked quite well.

00:02:14.640 --> 00:02:16.800 align:middle line:84%
Of course, I had
some occlusion when

00:02:16.800 --> 00:02:21.000 align:middle line:84%
there's so much interaction
between the dancers,

00:02:21.000 --> 00:02:22.860 align:middle line:84%
but it actually
worked quite well.

00:02:22.860 --> 00:02:25.770 align:middle line:84%
And in addition to the markers
that are put on the joints

00:02:25.770 --> 00:02:27.540 align:middle line:84%
that I was actually
interested in,

00:02:27.540 --> 00:02:31.500 align:middle line:84%
I also had quite a
few control markers

00:02:31.500 --> 00:02:36.780 align:middle line:84%
so I could differentiate
between the two dancers.

00:02:36.780 --> 00:02:39.242 align:middle line:84%
And you've been doing
studies in the lab,

00:02:39.242 --> 00:02:41.700 align:middle line:84%
but you've also been doing some
studies outside of the lab.

00:02:41.700 --> 00:02:45.630 align:middle line:84%
And what do you feel are the
most important differences

00:02:45.630 --> 00:02:48.840 align:middle line:84%
between doing something
inside of the lab and outside?

00:02:48.840 --> 00:02:51.630 align:middle line:84%
Of course, in the lab
it's more controlled

00:02:51.630 --> 00:02:56.300 align:middle line:84%
and you have control over
the placement of the camera.

00:02:56.300 --> 00:02:58.260 align:middle line:84%
So, when you go
somewhere, especially

00:02:58.260 --> 00:03:02.020 align:middle line:84%
if you don't necessarily
know them, the room,

00:03:02.020 --> 00:03:05.670 align:middle line:84%
and how to place the cameras,
that could be challenging.

00:03:05.670 --> 00:03:10.770 align:middle line:84%
But it's also nice to be able
to go where the people are,

00:03:10.770 --> 00:03:14.640 align:middle line:84%
that they don't have
to come into the lab.

00:03:14.640 --> 00:03:18.000 align:middle line:84%
That you can actually
just visit them.

00:03:18.000 --> 00:03:22.500 align:middle line:84%
And maybe that also makes
them more comfortable.

00:03:22.500 --> 00:03:26.340 align:middle line:84%
And I remember once you
came back from a field trip

00:03:26.340 --> 00:03:29.250 align:middle line:84%
like this, then you also talked
about the floor, and some

00:03:29.250 --> 00:03:30.780 align:middle line:90%
of the challenges of the floor.

00:03:30.780 --> 00:03:31.800 align:middle line:90%
What was that about?

00:03:31.800 --> 00:03:33.780 align:middle line:90%
Yeah.

00:03:33.780 --> 00:03:37.530 align:middle line:84%
Because some floors are
actually like bouncing,

00:03:37.530 --> 00:03:39.570 align:middle line:90%
and I had tripods.

00:03:39.570 --> 00:03:42.390 align:middle line:84%
So I had to make sure
that the tripods were

00:03:42.390 --> 00:03:45.960 align:middle line:84%
far away from the dancers,
because if they were close

00:03:45.960 --> 00:03:49.890 align:middle line:84%
then the tripods
would start to sway.

00:03:49.890 --> 00:03:51.780 align:middle line:90%
And that's not good.

00:03:51.780 --> 00:03:53.490 align:middle line:84%
Because then
actually the cameras

00:03:53.490 --> 00:03:56.500 align:middle line:84%
that started moving,
not the people only.

00:03:56.500 --> 00:03:59.100 align:middle line:84%
But then I guess it would be
better to hang the cameras,

00:03:59.100 --> 00:04:00.780 align:middle line:84%
but it may not
have been possible.

00:04:00.780 --> 00:04:02.280 align:middle line:84%
That was not possible
in that case,

00:04:02.280 --> 00:04:08.820 align:middle line:84%
but I think that's the
best if that's possible.

00:04:08.820 --> 00:04:12.450 align:middle line:84%
And if you can show me
something on the screen here.

00:04:12.450 --> 00:04:15.040 align:middle line:90%
Let's have a look.

00:04:15.040 --> 00:04:19.019 align:middle line:84%
So there we have
some of the dancers.

00:04:19.019 --> 00:04:22.650 align:middle line:84%
Here is from the
recording that we

00:04:22.650 --> 00:04:29.010 align:middle line:84%
were talking about, with the one
fiddler and the dance couple.

00:04:29.010 --> 00:04:33.480 align:middle line:84%
And here in this
representation I actually

00:04:33.480 --> 00:04:37.260 align:middle line:84%
removed the control
markers, because it's

00:04:37.260 --> 00:04:40.320 align:middle line:84%
for identifying the markers
but I don't need them.

00:04:40.320 --> 00:04:43.620 align:middle line:84%
So control marker, what
do you mean by that?

00:04:43.620 --> 00:04:51.930 align:middle line:84%
It's, for example, in
this dance they hold them,

00:04:51.930 --> 00:04:54.900 align:middle line:84%
they have their arms
like this all the time.

00:04:54.900 --> 00:04:59.190 align:middle line:84%
And I need to know if
this is the female dancer,

00:04:59.190 --> 00:05:01.470 align:middle line:90%
or the male dancer.

00:05:01.470 --> 00:05:03.750 align:middle line:84%
So, then I make sure
that the distance

00:05:03.750 --> 00:05:07.590 align:middle line:84%
between the joint markers
and the control markers,

00:05:07.590 --> 00:05:10.230 align:middle line:84%
that is between these
markers, that they

00:05:10.230 --> 00:05:13.740 align:middle line:90%
are different in the two arms.

00:05:13.740 --> 00:05:17.740 align:middle line:84%
You can use that afterwards then
to identify which one is which.

00:05:17.740 --> 00:05:19.590 align:middle line:84%
But you don't use it
then in the analysis.

00:05:19.590 --> 00:05:20.190 align:middle line:90%
No.

00:05:20.190 --> 00:05:21.750 align:middle line:90%
It's only for making that model.

00:05:21.750 --> 00:05:22.602 align:middle line:90%
Yes.

00:05:22.602 --> 00:05:24.060 align:middle line:84%
Are there any other
tricks that you

00:05:24.060 --> 00:05:27.690 align:middle line:84%
have learned from doing
these type of studies?

00:05:27.690 --> 00:05:31.530 align:middle line:90%
Well, it's the camera placement.

00:05:31.530 --> 00:05:34.230 align:middle line:84%
As I said the dancers
are moving quite about,

00:05:34.230 --> 00:05:36.730 align:middle line:84%
but the fiddler
is sitting still.

00:05:36.730 --> 00:05:41.700 align:middle line:84%
So, I had some cameras that
was only for the fiddler,

00:05:41.700 --> 00:05:46.410 align:middle line:84%
and then I had more
cameras for the dancers.

00:05:46.410 --> 00:05:49.410 align:middle line:84%
So, that is also a
trick to kind of make

00:05:49.410 --> 00:05:53.220 align:middle line:90%
a separate box for the fiddler.

00:05:53.220 --> 00:05:54.990 align:middle line:84%
And what about the
sound recording

00:05:54.990 --> 00:05:56.680 align:middle line:90%
during a setup like this?

00:05:56.680 --> 00:05:58.860 align:middle line:90%
How did you handle that?

00:05:58.860 --> 00:06:02.700 align:middle line:84%
Yeah, so I did a
separate sound recording.

00:06:02.700 --> 00:06:06.080 align:middle line:84%
That can be a bit
challenging, and it's more

00:06:06.080 --> 00:06:08.970 align:middle line:90%
to do some synchronisation.

00:06:08.970 --> 00:06:12.780 align:middle line:84%
And how did you do the
synchronisation in there?

00:06:12.780 --> 00:06:16.170 align:middle line:84%
I actually used a
clapboard with markers on.

00:06:16.170 --> 00:06:17.190 align:middle line:90%
Yeah.

00:06:17.190 --> 00:06:19.440 align:middle line:84%
Because the nice thing with
then having the markers on

00:06:19.440 --> 00:06:21.398 align:middle line:84%
is that you can see then
the clap in the motion

00:06:21.398 --> 00:06:24.300 align:middle line:84%
capture as well,
and get that spike.

00:06:24.300 --> 00:06:25.440 align:middle line:90%
Exactly.

00:06:25.440 --> 00:06:28.470 align:middle line:90%
That's a smart way of doing it.

00:06:28.470 --> 00:06:31.860 align:middle line:84%
Cool, so then you see there
are many different things

00:06:31.860 --> 00:06:34.110 align:middle line:84%
to learn from this
kind of use cases

00:06:34.110 --> 00:06:36.720 align:middle line:84%
and also take a look at
some of Mari's papers

00:06:36.720 --> 00:06:41.120 align:middle line:84%
as well that I'll link up
here in the text below.

00:06:41.120 --> 00:06:51.456 align:middle line:90%