All language subtitles for 04. Smoothing Tracking Data

af Afrikaans
ak Akan
sq Albanian
am Amharic
ar Arabic Download
hy Armenian
az Azerbaijani
eu Basque
be Belarusian
bem Bemba
bn Bengali
bh Bihari
bs Bosnian
br Breton
bg Bulgarian
km Cambodian
ca Catalan
ceb Cebuano
chr Cherokee
ny Chichewa
zh-CN Chinese (Simplified)
zh-TW Chinese (Traditional)
co Corsican
hr Croatian
cs Czech
da Danish
nl Dutch
en English
eo Esperanto
et Estonian
ee Ewe
fo Faroese
tl Filipino
fi Finnish
fr French
fy Frisian
gaa Ga
gl Galician
ka Georgian
de German
el Greek
gn Guarani
gu Gujarati
ht Haitian Creole
ha Hausa
haw Hawaiian
iw Hebrew
hi Hindi
hmn Hmong
hu Hungarian
is Icelandic
ig Igbo
id Indonesian
ia Interlingua
ga Irish
it Italian
ja Japanese
jw Javanese
kn Kannada
kk Kazakh
rw Kinyarwanda
rn Kirundi
kg Kongo
ko Korean
kri Krio (Sierra Leone)
ku Kurdish
ckb Kurdish (Soranî)
ky Kyrgyz
lo Laothian
la Latin
lv Latvian
ln Lingala
lt Lithuanian
loz Lozi
lg Luganda
ach Luo
lb Luxembourgish
mk Macedonian
mg Malagasy
ms Malay
ml Malayalam
mt Maltese
mi Maori
mr Marathi
mfe Mauritian Creole
mo Moldavian
mn Mongolian
my Myanmar (Burmese)
sr-ME Montenegrin
ne Nepali
pcm Nigerian Pidgin
nso Northern Sotho
no Norwegian
nn Norwegian (Nynorsk)
oc Occitan
or Oriya
om Oromo
ps Pashto
fa Persian
pl Polish
pt-BR Portuguese (Brazil)
pt Portuguese (Portugal)
pa Punjabi
qu Quechua
ro Romanian
rm Romansh
nyn Runyakitara
ru Russian
sm Samoan
gd Scots Gaelic
sr Serbian
sh Serbo-Croatian
st Sesotho
tn Setswana
crs Seychellois Creole
sn Shona
sd Sindhi
si Sinhalese
sk Slovak
sl Slovenian
so Somali
es Spanish
es-419 Spanish (Latin American)
su Sundanese
sw Swahili
sv Swedish
tg Tajik
ta Tamil
tt Tatar
te Telugu
th Thai
ti Tigrinya
to Tonga
lua Tshiluba
tum Tumbuka
tr Turkish
tk Turkmen
tw Twi
ug Uighur
uk Ukrainian
ur Urdu
uz Uzbek
vi Vietnamese
cy Welsh
wo Wolof
xh Xhosa
yi Yiddish
yo Yoruba
zu Zulu

Original subtitles

In this lesson,

we are going to learn a few different ways that we can

affect our tracking data to get a smoother,

less noisy track.

OK, so I have this scene set up for you already,

and we are just starting out in your project files in 04 underscore begin.

And I am going to go ahead and just play this through so

you can see what is going on here.

So what I have tracked in this scene is this watch on our guy's hand.

And, originally, that watch was orange.

So I have tracked the watch and added a roto shape just to outline that watch.

And then I have added a color correction to desaturate it.

So if I select the color correction node and we hit D,

just to disable that node, and you see it gets that little x through it.

And the color correction goes away, and we see that original color.

Now this is something that you might have to use tracking

for to desaturate something over time.

But that is not what this lesson is about.

This lesson is about smoothing out your track.

So I did not want you to take any time just to

figure out how to set all of this up.

It is already done for you.

So, basically,

this type of shot is notorious for having a noisy or jittery track.

So let us watch that play through again.

What is happening is we have a really smooth camera pan that is

noticeably smooth as it moves across our scene.

So you can see it just very smoothly goes from one side to the other.

We do not get any kind of a rotation up and down.

It does not feel handheld.

This is shot on a dolly.

So these kinds of shots are notorious for having noisy tracks

because they themselves are so obviously very smooth.

So, in the end,

really what happens is the camera itself might move around just a little bit.

Maybe you get something, like even carpet fibers,

whenever you are pushing that dolly across the floor can

give you just a little bit of bounce.

And to the eye, the whole track still looks very smooth.

Or not the track, the shot looks very smooth.

But then whenever you start using tracking data from that shot,

it starts to have a little bit of a bounce in it,

a little bit of noise.

So I want to show you how to fix that.

And definitely a pan like this is not going to be the

only time that you see noise in a shot, but it is definitely one of the worst.

So let us go ahead and select our tracker.

I am just going to double-click that so you can see what we have here.

And then go over to your Tracker tab.

So I want you to just take a look at what we have got here.

So remember how I said that shot was very smooth.

And now what we are seeing here is not a straight line.

We get this little up and down motion with our different points.

So this happens partially because the tracker itself is not 100% accurate.

As things move across the screen, they are going to change pixel color,

and that is what the tracker has to use.

So it is not always going to be exactly the same.

And you get these tiny little variations.

Now this is not bad noise.

You are going to see probably a lot worse noise as you learn how

to track more and you run into more things.

But the way that we fix this is definitely a balancing act.

So I am going to show you how to go ahead and get started with that.

Let us come over here to our transform tab in our Tracker node.

And you see that we have this option here, the Smooth.

So then we also have the T, R, and S.

So what that is saying is the amount we want to smooth the translation,

the rotation, and the scale.

Now,

I do not want to change the scale of rotation because this

was simply a translation-style track, up and down,

left and right info.

So what I want to do is to come into this box here and we will

just enter in the value 1 and hit Enter.

Now you can see that we get a little bit of a difference in our lines.

Now, originally it was kind of hard to see,

but that red line was underneath the gray line.

And this gray line now is a little bit more straight.

You can see the red line goes up and down behind that gray line.

So even with a value of 1, we are getting a lot straighter,

more just smooth track.

Now, a lot of the time,

I would not recommend changing this a lot more than a 1 if you are

trying to do something like what I have done here,

where you are tracking a specific object that you are trying to cover up.

Now,

if you wanted to just get tracking data and then have a

completely different object move along with that,

then you could get away with more smoothing.

But notice if I come into this smooth box and,

let us say, turn this up to a value of something like 10,

even though the track is more smooth, the shape is not staying in place.

So what I meant by balancing act earlier was just that,

if you smooth it too much, you are not going to have an accurate track.

And you are basically destroying the purpose of tracking in the first place.

But you can just use a smaller value like a 1,

maybe a 2,

and that is just going to give you that smoother track and just make

that watch appear to be a little-- or the mask around the watch to be

a little more integrated into your shot.

Now,

another way that you can fix tracking data before you have ever

even gotten to the point where we are at now,

where you have already tracked this is in the Settings tab.

So if you click that Settings tab, we have got a pre-track filter.

Now, in NUKE 7.

0, this defaults to adjust contrast And so,

basically, what that is doing is it is calculating,

if it had a color correction on the whole shot and we bumped up those

whites and pushed up those blacks to make it more contrasty,

it is going to have a little bit more accurate track.

And we are going to talk about how to actually manually add a color correction

to help us get better tracking data later on in this course.

But it can do this in the background just with this setting chosen here.

But we also have this other setting that is the median.

And choosing median before you do a track is going to help you to get better,

less noisy data as well.

It can take a little bit longer to process.

But if you are trying to track something, it is probably worth it.

So you might just experiment between using the median and

the adjust contrast before you do your track to see what

gets you the least jittery track.

And sometimes median is not going to help you.

It is a very shot by shot basis.

But most of the time I really prefer to use the smooth.

I get pretty much instant results,

and I do not have to redo tracks if something did not work for me.

So those are just a couple ways that you can work with smoothing out your data.

Now that we have learned how to smooth out our data and we have also,

in our previous lessons, learned how to do a 1-point track and apply that data,

we are going to move on to learning how to do a 2-point track.

So that is going to require a little bit more finesse over here in the tracker,

a few more settings.

But it is not that difficult.

So stick around.

We are going to learn how to do that.

And why you would use a 1-point track versus a

2-point track in certain situations?

And we are going to learn all of that in our next lesson.

Can't find what you're looking for?
Get subtitles in any language from opensubtitles.com, and translate them here.