TEDxUSC
Paul Debevec: Animating a photo-real digital face
Paul Debevec animira foto-realistično digitalno lice
Filmed:
Readability: 4.8
881,584 views
Na događaju TEDxUSC, utirač puteva u računalnoj grafici Paul Debevec objašnjava tehnologiju snimanja scene koja je u pozadini digitalne Emily, digitalno sklopljenog ljudskog lica tako realističnog da podnosi i višestruke provjere.
Paul Debevec - Computer graphics pioneer
Paul Debevec's digital inventions have powered the breathtaking visual effects in films like The Matrix, Superman Returns, King Kong and The Curious Case of Benjamin Button. Full bio
Paul Debevec's digital inventions have powered the breathtaking visual effects in films like The Matrix, Superman Returns, King Kong and The Curious Case of Benjamin Button. Full bio
Double-click the English transcript below to play the video.
00:18
One of the biggest challenges in computer graphics
0
0
3000
Jedan od najvećih izazova u računalnoj grafici
00:21
has been being able to create a photo-real,
1
3000
2000
je biti u stanju stvoriti foto-realistično,
00:23
digital human face.
2
5000
2000
digitalno ljudsko lice.
00:25
And one of the reasons it is so difficult is that, unlike aliens and dinosaurs,
3
7000
3000
A jedan od razloga zašto je to teško je taj što, za razliku od dinosaura,
00:28
we look at human faces every day.
4
10000
3000
ljudska lica promatramo svakodnevno.
00:31
They are very important to how we communicate with each other.
5
13000
2000
Vrlo su važna za našu međusobnu komunikaciju.
00:33
As a result, we're tuned in to the subtlest things
6
15000
3000
Kao rezultat, primjećujemo najtananije stvari
00:36
that could possibly be wrong with a computer rendering,
7
18000
2000
koje bi uopće mogle biti pogrešne u računalnom renderiranju,
00:38
in order to believe whether these things are realistic.
8
20000
3000
kako bismo vjerovali jesu li te stvari realistične.
00:41
And what I'm going to do in the next five minutes
9
23000
2000
Ono što ću učiniti u sljedećih pet minuta
00:43
is take you through a process
10
25000
2000
je da ću vas provesti kroz proces
00:45
where we tried to create a reasonably photo-realistic computer-generated face,
11
27000
3000
kojim smo pokušali stvoriti razumno foto-realistično računalno generirano lice,
00:48
using some computer graphics technology we've developed,
12
30000
3000
koristeći tehnologiju računalne grafike koju smo sami razvili,
00:51
and also some collaborators at a company called Image Metrics.
13
33000
2000
kao i suradnike u kompaniji po imenu Image Metrics.
00:53
And we're going to try to do a photo-real face
14
35000
2000
Pokušat ćemo napraviti foto-realistično lice
00:55
of an actress named Emily O'Brien, who is right there.
15
37000
3000
glumice po imenu Emily O'Brien, koja je ovdje.
00:58
And that's actually a completely computer-generated rendering of her face.
16
40000
3000
A to je zapravo potpuno računalno renderiranje njezinog lica.
01:01
By the end of the talk, we're going to see it move.
17
43000
2000
Do kraja ovog govora, vidjet ćemo ga u pokretu.
01:03
The way that we did this is we tried to start with Emily herself,
18
45000
3000
Način na koji smo to napravili je tako da smo pokušali početi sa samom Emily,
01:06
who was gracious enough to come to our laboratory
19
48000
3000
koja je bila dovoljno draga i došla do našeg laboratorija
01:09
in Marina Del Rey, and sit for a session in Light Stage 5.
20
51000
3000
u marini Del Rey, i sjela na seansu na svjetlosnoj pozornici 5.
01:12
This is a face-scanning sphere, with 156 white LEDs all around
21
54000
4000
Ovo je kugla za skeniranje lica koja ima 156 bijelih LED-ova uokolo,
01:16
that allow us to photograph her
22
58000
2000
koji nam omogućuju da je fotografiramo
01:18
in a series of very controlled illumination conditions.
23
60000
3000
u nizu kontroliranih uvjeta osvjetljenja.
01:21
And the lighting that we use these days looks something like this.
24
63000
3000
A svjetlost koju koristimo ovih dana izgleda poput ovoga.
01:24
We shoot all of these photographs in about three seconds.
25
66000
2000
Slikamo sve ove fotografije u neke tri sekunde.
01:26
And we basically capture enough information
26
68000
3000
I praktički dobivamo dovoljno informacija
01:29
with video projector patterns that drape over the contours of her face,
27
71000
4000
s uzorcima video projekcije koji se prevlače preko njezinog lica,
01:33
and different principle directions of light from the light stage,
28
75000
2000
i različitim glavnim pravcima svjetla u svjetlosnoj pozornici,
01:35
to figure out both the coarse-scale
29
77000
2000
da raščlanimo i grube crte
01:37
and the fine-scale detail of her face.
30
79000
2000
kao i fine sitne detalje njezinog lica.
01:39
If we zoom in on this photograph right here,
31
81000
2000
Ako zumiramo ovu fotografiju ovdje,
01:41
we can see it's a really nice photograph to have of her,
32
83000
2000
vidimo da je to zaista lijepa fotografija nje,
01:43
because she is lit from absolutely everywhere at the same time
33
85000
4000
jer je osvijetljena s apsolutno svih strana u isto vrijeme
01:47
to get a nice image of her facial texture.
34
89000
2000
čime dobivamo lijepu sliku površine njezinog lica.
01:49
And in addition, we've actually used polarizers on all the lights --
35
91000
3000
Dodatno, koristili smo polarizatore na svim svjetlima --
01:52
just like polarized sunglasses can block
36
94000
2000
baš kao što polarizirane sunčane naočale blokiraju
01:54
the glare off of the road,
37
96000
2000
sjaj koji se reflektira od ceste,
01:56
polarizers can block the shine off of the skin,
38
98000
2000
polarizatori mogu blokirati sjaj kože,
01:58
so we don't get all those specular reflections to take this map.
39
100000
3000
pa ne dobivamo refleksije na ovoj mapi.
02:01
Now, if we turn the polarizers around just a little bit,
40
103000
2000
No, ako okrenemo polarizatore samo malo,
02:03
we can actually bring that specular reflection
41
105000
2000
možemo zapravo tu refleksiju
02:05
of the skin back in,
42
107000
2000
vratiti na lice,
02:07
and you can see she looks kind of shiny and oily at this point.
43
109000
3000
i vidite da ona sada izgleda nekako sjajno i nauljeno.
02:10
If you take the difference between these two images here,
44
112000
3000
Ako uzmete razliku između ove dvije slike,
02:13
you can get an image lit from the entire sphere of light
45
115000
2000
možete dobiti sliku osvijetljenu s cjelokupnom svjetlosnom kuglom
02:15
of just the shine off of Emily's skin.
46
117000
3000
ili samo sjaj s Emilyne kože.
02:18
I don't think any photograph like this had ever been taken
47
120000
2000
Mislim da fotografija poput ove nikada nije bila fotografirana
02:20
before we had done this.
48
122000
2000
prije nego što smo mi ovo učinili.
02:22
And this is very important light to capture,
49
124000
3000
A ovo je vrlo važno svjetlo za uhvatiti,
02:25
because this is the light that reflects off the first surface of the skin.
50
127000
2000
jer je to svjetlo koje se reflektira s prve površine kože.
02:27
It doesn't get underneath the translucent
51
129000
2000
Ono ne ulazi pod prozirne
02:29
layers of the skin and blur out.
52
131000
2000
slojeve kože i tu se muti.
02:31
And, as a result, it's a very good cue
53
133000
2000
Rezultat je taj da imamo vrlo dobar pokazatelj
02:33
to the detailed shape of the skin-pore structure
54
135000
3000
detaljnog oblika strukture kože i pora
02:36
and all of the fine wrinkles that all of us have,
55
138000
2000
i svih finih bora koje svi imamo,
02:38
the things that actually make us look like real humans.
56
140000
2000
stvari zbog kojih zapravo i izgledamo kao pravi ljudi.
02:40
So, if we use information that comes off of this specular reflection,
57
142000
4000
Dakle, ako koristimo informacije koje dobivamo od te točkaste refleksije
02:44
we can go from a traditional face scan
58
146000
2000
možemo ići od tradicionalnog skena lica
02:46
that might have the gross contours of the face and the basic shape,
59
148000
4000
koji nam daje grube konture lica i glavni oblik,
02:50
and augment it with information
60
152000
2000
i pojačati ga s informacijama
02:52
that puts in all of that skin pore structure and fine wrinkles.
61
154000
3000
koje dodaju čitavu strukturu pora i fine bore.
02:55
And, even more importantly,
62
157000
2000
I još važnije,
02:57
since this is a photometric process that only takes three seconds to capture,
63
159000
3000
otkad je to fotometrijski proces koji zahtijeva samo 3 sekunde,
03:00
we can shoot Emily
64
162000
2000
možemo fotografirati Emily
03:02
in just part of an afternoon,
65
164000
2000
tijekom samo dijela popodneva,
03:04
in many different facial poses and facial expressions.
66
166000
3000
u mnogo različitih položaja lica i mnogo izraza.
03:07
So, here you can see her moving her eyes around, moving her mouth around.
67
169000
3000
Ovdje je vidite kako okreće očima, pomiče usta.
03:10
And these we're actually going to use to create a photo-real digital character.
68
172000
3000
A ove ćemo zaista koristiti da stvorimo foto-realističan digitalni karakter.
03:13
If you take a look at these scans that we have of Emily,
69
175000
4000
Ako pogledate ove skenove koje imamo od Emily,
03:17
you can see that the human face does an enormous amount of amazing things
70
179000
3000
vidite da ljudsko lice čini ogroman broj zapanjujućih stvari
03:20
as it goes into different facial expressions.
71
182000
3000
kad pravi različite izraze.
03:23
You can see things. Not only the face shape changes,
72
185000
3000
Možete to vidjeti. Ne samo da se oblik lica mijenja,
03:26
but all sorts of different skin buckling and skin wrinkling occurs.
73
188000
4000
već se događa i različito savijanje i nabiranje kože.
03:30
You can see that the skin pore structure changes enormously
74
192000
4000
Vidite da se struktura pora kože drastično mijenja
03:34
from stretched skin pores
75
196000
2000
od rastegnutih pora,
03:36
to the regular skin texture.
76
198000
2000
do normalne teksture kože.
03:38
You can see the furrows in the brow and how the microstructure changes there.
77
200000
3000
Možete vidjeti brazde na čelu i kako se mikrostruktura ovdje mijenja.
03:41
You can see muscles pulling down at flesh to bring her eyebrows down.
78
203000
4000
Vidite kako mišići povlače kožu nadolje kako bi spustili njezine obrve.
03:45
Her muscles bulging in her forehead when she winces like that.
79
207000
3000
Mišići iskaču na njezinom čelu kad pravi ovu grimasu.
03:48
In addition to this kind of high-resolution geometry,
80
210000
3000
Osim ove geometrije visoke rezolucije,
03:51
since it's all captured with cameras, we've got a great texture map to use for the face.
81
213000
4000
budući da je sve snimljeno fotoaparatom, imamo odličnu mapu teksture za upotrijebiti na licu.
03:55
And by looking at how the different color channels of the illumination,
82
217000
2000
A gledajući kako različiti kanali boja u svjetlosti,
03:57
the red and the green and the blue,
83
219000
2000
crveni i zeleni i plavi,
03:59
diffuse the light differently,
84
221000
2000
različito raspršuju svjetlo,
04:01
we can come up with a way of shading the skin on the computer.
85
223000
3000
možemo odrediti način na koji ćemo osjenčati kožu računalom.
04:04
Then, instead of looking like a plaster mannequin,
86
226000
3000
I tada, umjesto da izgleda kao gipsana lutka,
04:07
it actually looks like it's made out of living human flesh.
87
229000
3000
ovo zapravo izgleda kao da je izrađeno od živog ljudskog tkiva.
04:10
And this is what we used
88
232000
2000
I to je ono što smo davali
04:12
to give to the company Image Metrics
89
234000
2000
kompaniji Image Metrics
04:14
to create a rigged, digital version of Emily.
90
236000
3000
da naprave prilagođenu digitalnu verziju Emily.
04:17
We're just seeing the coarse-scale geometry here.
91
239000
2000
Ovdje vidimo samo geometriju u grubom planu.
04:19
But they basically created a digital puppet of her,
92
241000
3000
Ali oni su napravili digitalnu marionetu nje,
04:22
where you can pull on these various strings,
93
244000
2000
kod koje možete povući ove različite uzice,
04:24
and it actually moves her face in ways that are
94
246000
2000
i one pomiču njezino lice na način
04:26
completely consistent with the scans that we took.
95
248000
2000
sasvim konzistentan sa skenovima koje smo napravili.
04:28
And, in addition to the coarse-scale geometry,
96
250000
2000
Osim grube geometrije,
04:30
they also used all of that detail
97
252000
2000
Iskoristili su i sve te detalje
04:32
to create a set of what are called "displacement maps"
98
254000
3000
da naprave skup onoga što se zove "karte pomaka"
04:35
that animate as well.
99
257000
3000
koje se također mogu animirati.
04:38
These are the displacement maps here.
100
260000
2000
Ovo su karte pomaka.
04:40
And you can see those different wrinkles actually show up as she animates.
101
262000
3000
I vidite da se različite bore doista vide kad je animirana.
04:43
So the next process was then to animate her.
102
265000
2000
Sljedeći je postupak, dakle, bila animacija.
04:45
We actually used one of her own performances to provide the source data.
103
267000
4000
Koristili smo jednu njenu pravu izvedbu kao izvor podataka.
04:49
So, by analyzing this video with computer vision techniques,
104
271000
3000
Analizirajući ovaj video tehnikama računalnog gledanja,
04:52
they were able to drive the facial rig
105
274000
2000
bili su u stanju pokrenuti kostur lica
04:54
with the computer-generated performance.
106
276000
2000
izvedbom generiranom u računalu.
04:56
So what you're going to see now, after this,
107
278000
2000
Ono što ćete sada vidjeti, nakon ovoga,
04:58
is a completely photo-real digital face.
108
280000
2000
je posve foto-realistično lice.
05:00
We can turn the volume up a little bit if that's available.
109
282000
2000
Možemo malo pojačati glasnoću ako može.
05:02
Emily: Image Metrics is a markerless, performance-driven animation company.
110
284000
4000
Emily: Image Metrics je kompanija za animaciju bez markera, temeljenu na izvedbi.
05:06
We specialize in high-quality facial animation
111
288000
3000
Specijaliziramo se za animaciju lica visoke kvalitete
05:09
for video games and films.
112
291000
2000
za video igre i filmove.
05:11
Image Metrics is a markerless, performance-driven animation company.
113
293000
4000
Image Metrics je kompanija za animaciju bez markera, temeljenu na izvedbi.
05:15
We specialize in high quality facial animation
114
297000
3000
Specijaliziramo se za animaciju lica visoke kvalitete
05:18
for video games and films.
115
300000
3000
za video igre i filmove.
05:21
Paul Debevec: So, if we break that down into layers, here's that diffuse component
116
303000
2000
Paul Debevec: Pa, ako to raščlanimo na slojeve, evo te difuzne komponente
05:23
we saw in the first slide.
117
305000
2000
koju smo vidjeli na prvom slajdu.
05:25
Here is the specular component animating.
118
307000
2000
Ovdje je refleksivna komponenta u animaciji.
05:27
You can see all the wrinkles happening there.
119
309000
2000
Vidite da se ovdje pojavljuju i sve bore.
05:29
And there is the underlying wireframe mesh.
120
311000
5000
A ovdje je žičana mreža u podlozi.
05:34
And that is Emily herself.
121
316000
2000
A ovo je sama Emily.
05:36
Now, where are we going with this here?
122
318000
2000
Pa, kamo dalje s ovime?
05:38
We've gone a little bit beyond Light Stage 5. This is Light Stage 6,
123
320000
4000
Pomaknuli smo se nešto dalje od svjetlosne pozornice 5. Ovo je svjetlosna pozornica 6.
05:42
and we're looking at taking this technology
124
324000
2000
I gledamo kako iskoristiti ovu tehnologiju
05:44
and applying it to whole human bodies.
125
326000
2000
i primijeniti je na čitava tijela.
05:46
This is Bruce Lawmen, one of our researchers in the group,
126
328000
2000
Ovo je Bruce Lawmen, jedan od istraživača u skupini,
05:48
who graciously agreed to get captured running in the Light Stage.
127
330000
3000
koji se velikodušno složio da ga snimimo dok trči na svjetlosnoj pozornici.
05:51
And let's take a look at a computer-generated version
128
333000
2000
Pogledajmo verziju Brucea generiranu računalom,
05:53
of Bruce, running in a new environment.
129
335000
2000
dok trči u novom okruženju.
06:01
And thank you very much.
130
343000
2000
Hvala vam lijepa.
06:03
(Applause)
131
345000
1000
(Pljesak)
ABOUT THE SPEAKER
Paul Debevec - Computer graphics pioneerPaul Debevec's digital inventions have powered the breathtaking visual effects in films like The Matrix, Superman Returns, King Kong and The Curious Case of Benjamin Button.
Why you should listen
Called "Hollywood's Master of Light" by MIT's Technology Review, Paul Debevec leads the Graphics Laboratory at USC's Institute for Creative Technologies, where he directs the development of the Light Stage systems, which capture and simulate how people and objects appear in real-world illumination.
His work on lighting digital objects has been used in films like X-Men, The Golden Compass and Terminator: Salvation. He leads the design of HDR Shop, the first high-dynamic-range image editing program. He has collaborated with TEDster Golan Levin on an interactive art installation called Rouen Revisited.
Paul Debevec | Speaker | TED.com