All language subtitles for 03 - Course Introduction and Prepping Our Plate - De-noising and Adding Back the Real Grain

af Afrikaans
sq Albanian
am Amharic
ar Arabic Download
hy Armenian
az Azerbaijani
eu Basque
be Belarusian
bn Bengali
bs Bosnian
bg Bulgarian
ca Catalan
ceb Cebuano
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
tl Filipino
fi Finnish
fr French
fy Frisian
gl Galician
ka Georgian
de German
el Greek
gu Gujarati
ht Haitian Creole
ha Hausa
haw Hawaiian
iw Hebrew
hi Hindi
hmn Hmong
hu Hungarian
is Icelandic
ig Igbo
id Indonesian
ga Irish
it Italian
ja Japanese
jw Javanese
kn Kannada
kk Kazakh
km Khmer
ko Korean
ku Kurdish (Kurmanji)
ky Kyrgyz
lo Lao
la Latin
lv Latvian
lt Lithuanian
lb Luxembourgish
mk Macedonian
mg Malagasy
ms Malay
ml Malayalam
mt Maltese
mi Maori
mr Marathi
mn Mongolian
my Myanmar (Burmese)
ne Nepali
no Norwegian
ps Pashto
fa Persian
pl Polish
pt Portuguese
pa Punjabi
ro Romanian
ru Russian
sm Samoan
gd Scots Gaelic
sr Serbian
st Sesotho
sn Shona
sd Sindhi
si Sinhala
sk Slovak
sl Slovenian
so Somali
es Spanish
su Sundanese
sw Swahili
sv Swedish
tg Tajik
ta Tamil
te Telugu
th Thai
tr Turkish
uk Ukrainian
ur Urdu
uz Uzbek
vi Vietnamese
cy Welsh
xh Xhosa
yi Yiddish
yo Yoruba
zu Zulu
or Odia (Oriya)
rw Kinyarwanda
tk Turkmen
tt Tatar
ug Uyghur
Would you like to inspect the original subtitles? These are the user uploaded subtitles that are being translated: 1 00:00:01,066 --> 00:00:02,199 so in this clip 2 00:00:02,200 --> 00:00:05,033 we gonna start off with prepping the plate 3 00:00:05,300 --> 00:00:06,266 just so you know 4 00:00:06,266 --> 00:00:08,466 I have flipped the image 5 00:00:08,500 --> 00:00:11,233 this is how it was originally shot 6 00:00:11,700 --> 00:00:14,966 with her looking to the right and not left as 7 00:00:15,200 --> 00:00:16,966 as it was in the end video 8 00:00:17,500 --> 00:00:19,000 and before we do anything 9 00:00:19,000 --> 00:00:21,700 we gonna set up the project settings 10 00:00:21,700 --> 00:00:23,400 so the source of our shot 11 00:00:23,400 --> 00:00:27,766 it is shot at half HD so 720 p 12 00:00:28,733 --> 00:00:31,666 and we gonna make sure our project 13 00:00:31,666 --> 00:00:33,399 is set up correctly here 14 00:00:35,266 --> 00:00:38,199 so we're gonna set it to half HD 15 00:00:38,800 --> 00:00:41,766 so before I start doing any composting 16 00:00:41,766 --> 00:00:44,133 there's a few tasks that I like to do 17 00:00:44,133 --> 00:00:46,566 when I approach a new plate 18 00:00:47,000 --> 00:00:49,800 so I always start with running a 19 00:00:49,800 --> 00:00:52,566 d noise is usually number one 20 00:00:53,500 --> 00:00:56,100 and then do my tracking 21 00:00:56,100 --> 00:00:57,900 as we will do in the next clip 22 00:00:57,900 --> 00:01:00,466 because these two tasks are kind of needed 23 00:01:00,466 --> 00:01:02,699 in order to do any of the other stuff 24 00:01:02,700 --> 00:01:04,500 we will be doing next 25 00:01:04,800 --> 00:01:08,500 so I like to prepare and finish all of that first 26 00:01:08,600 --> 00:01:11,633 so you can focus on the other stuff later 27 00:01:12,166 --> 00:01:14,333 the important thing about d noise 28 00:01:14,333 --> 00:01:17,999 is that we needed to do a better key 29 00:01:18,000 --> 00:01:19,366 to pull a better key on this 30 00:01:19,366 --> 00:01:22,299 and if you look at this footage 31 00:01:22,533 --> 00:01:26,066 it's not really a green screen right 32 00:01:26,066 --> 00:01:27,533 it's just a wall 33 00:01:27,533 --> 00:01:28,799 you know at 34 00:01:28,800 --> 00:01:31,866 at first I didn't shoot this with the intention 35 00:01:31,866 --> 00:01:33,766 to actually key the background 36 00:01:34,066 --> 00:01:35,333 but then I figured 37 00:01:35,333 --> 00:01:36,866 like why not 38 00:01:36,866 --> 00:01:41,166 so I ended up pulling a bit of a luminance key on this 39 00:01:41,166 --> 00:01:45,799 and then doing a few tricks to easily get this hair in 40 00:01:46,700 --> 00:01:49,366 because a lot of people know from experience that 41 00:01:49,600 --> 00:01:52,533 keying hair is really difficult 42 00:01:52,533 --> 00:01:57,866 and it's very difficult to get these fine details 43 00:01:58,200 --> 00:02:00,400 but in many times 44 00:02:00,866 --> 00:02:04,166 you can avoid that using some other simple tricks if 45 00:02:04,166 --> 00:02:06,166 if the footage look good that is 46 00:02:06,400 --> 00:02:10,866 it's important that we do a good d noise 47 00:02:11,133 --> 00:02:12,366 or a decent d noise 48 00:02:12,366 --> 00:02:15,333 just so we can isolate 49 00:02:15,333 --> 00:02:18,033 the noise and the grain from the image 50 00:02:18,366 --> 00:02:21,899 because when we do any manipulations on the image 51 00:02:22,100 --> 00:02:25,200 we want to keep the grain intact 52 00:02:25,200 --> 00:02:28,566 you know we want to be able to remove the grain 53 00:02:28,966 --> 00:02:30,499 do our operations 54 00:02:30,500 --> 00:02:32,966 and then add it back on in the end 55 00:02:33,066 --> 00:02:34,966 to make it seem like we 56 00:02:35,066 --> 00:02:37,499 did not touch it in the beginning you know 57 00:02:37,600 --> 00:02:40,533 or you remove the grain completely 58 00:02:40,533 --> 00:02:41,599 and in the end 59 00:02:41,600 --> 00:02:42,400 you put like 60 00:02:42,400 --> 00:02:45,000 a universal grain 61 00:02:45,500 --> 00:02:48,366 that you put on all the shots and all the CG shots 62 00:02:48,366 --> 00:02:49,766 to make them match 63 00:02:50,066 --> 00:02:55,299 and this is usually a grain plate that was shot at set 64 00:02:55,300 --> 00:02:58,100 you know they shot a small clip on 65 00:02:58,100 --> 00:03:00,000 like a blank background 66 00:03:00,000 --> 00:03:01,800 just to isolate the grain 67 00:03:02,366 --> 00:03:07,399 okay so we won't really do anything complicated here 68 00:03:07,400 --> 00:03:08,400 when it comes to the grain 69 00:03:08,400 --> 00:03:11,233 you can see that this material is 70 00:03:11,300 --> 00:03:12,600 is not really good 71 00:03:12,600 --> 00:03:16,633 like this footage is not that good quality 72 00:03:18,333 --> 00:03:19,199 anyway let's 73 00:03:19,200 --> 00:03:21,933 let's start doing our d noise here 74 00:03:21,933 --> 00:03:25,799 and I usually just use the simple d noise included 75 00:03:26,733 --> 00:03:30,166 if you want a suggestion for a good denoys plugin 76 00:03:30,400 --> 00:03:33,033 you should look into neat video 77 00:03:33,333 --> 00:03:36,899 it's by far the best denoys plugin on the market 78 00:03:38,333 --> 00:03:42,099 I like to just look in some of the channels 79 00:03:42,100 --> 00:03:45,366 and the blue channel is usually the most crappy one 80 00:03:45,366 --> 00:03:48,099 as you can see with the grain wise so 81 00:03:49,500 --> 00:03:52,000 we just gonna make a basic one here 82 00:03:52,000 --> 00:03:53,166 I mean this digital 83 00:03:53,166 --> 00:03:55,899 so set it up to digital 84 00:03:55,900 --> 00:04:00,700 but it's always good to experiment with 85 00:04:01,466 --> 00:04:03,566 with these and see you know 86 00:04:03,733 --> 00:04:07,066 it's funny sometimes even though you short digital 87 00:04:07,366 --> 00:04:08,699 one of these settings 88 00:04:08,700 --> 00:04:10,700 like using film as a source 89 00:04:10,700 --> 00:04:12,266 might give you a better result 90 00:04:12,266 --> 00:04:13,099 you know it's 91 00:04:13,533 --> 00:04:16,699 it's all just testing 92 00:04:18,666 --> 00:04:21,133 okay and many people that say 93 00:04:21,133 --> 00:04:24,399 you gonna put this grain on top of your checkerboard 94 00:04:24,400 --> 00:04:30,000 right what many people do is they like to use the 95 00:04:31,200 --> 00:04:33,666 the furnace regrain right 96 00:04:33,933 --> 00:04:36,433 because that's the most commonly used 97 00:04:36,533 --> 00:04:38,099 basically this one 98 00:04:38,866 --> 00:04:43,566 tries to read the grain on a frame of the footage 99 00:04:44,166 --> 00:04:47,399 and then that's like a interpolation estimate 100 00:04:48,966 --> 00:04:50,499 so if we look here 101 00:04:55,200 --> 00:04:57,766 so you can see it add back the grain right 102 00:04:57,766 --> 00:04:59,499 but it's not the real grain 103 00:04:59,500 --> 00:05:01,200 you know that we see in the plate 104 00:05:01,200 --> 00:05:03,233 you can see that it differs some 105 00:05:04,100 --> 00:05:08,000 to get the absolute close match you can get 106 00:05:09,300 --> 00:05:10,500 what you want to do is 107 00:05:10,500 --> 00:05:14,566 you want to remove the grain from the image 108 00:05:14,933 --> 00:05:16,266 and isolate it 109 00:05:16,533 --> 00:05:21,399 and comping is all about simple image operations right 110 00:05:21,400 --> 00:05:22,700 so in the merge 111 00:05:24,333 --> 00:05:27,666 we have a few operations here and what I want I want to 112 00:05:27,666 --> 00:05:31,899 let's say I have my image without my noise right 113 00:05:32,333 --> 00:05:36,033 it means if I take my source image 114 00:05:36,533 --> 00:05:39,699 remove the image without the noise 115 00:05:39,866 --> 00:05:42,033 that should leave me only the noise 116 00:05:43,100 --> 00:05:48,133 so you can do this easy with a simple minus operation 117 00:05:48,133 --> 00:05:52,899 so a a minus B 118 00:05:54,133 --> 00:05:55,666 and you get the grain 119 00:05:56,666 --> 00:05:58,799 and a easy way to check this 120 00:05:58,800 --> 00:06:02,033 that everything goes as it should 121 00:06:02,366 --> 00:06:05,566 is you then can just do a plus 122 00:06:07,566 --> 00:06:09,866 back on your denoist 123 00:06:11,000 --> 00:06:16,000 and that should give you the original image you know 124 00:06:16,466 --> 00:06:19,466 because I you just isolating and removing 125 00:06:19,466 --> 00:06:22,533 and you can even see this 126 00:06:22,533 --> 00:06:26,499 you know I sample this and compare the values here 127 00:06:27,166 --> 00:06:28,633 there's no difference 128 00:06:29,300 --> 00:06:32,533 if I were to do like operation here 129 00:06:32,533 --> 00:06:33,799 let's put like 130 00:06:35,466 --> 00:06:37,299 multiply you know 131 00:06:37,566 --> 00:06:39,733 you could see that down here 132 00:06:39,733 --> 00:06:43,133 yeah we actually affect it's not 133 00:06:43,133 --> 00:06:44,466 it's not the same 134 00:06:45,500 --> 00:06:47,300 the original plate is not the same 135 00:06:47,300 --> 00:06:48,666 after we add back to grain 136 00:06:50,733 --> 00:06:51,666 but without 137 00:06:51,666 --> 00:06:53,499 it's identical you know 138 00:06:53,966 --> 00:06:55,866 so what you would do 139 00:06:56,466 --> 00:06:59,699 if your surface is flat in the background here 140 00:06:59,800 --> 00:07:01,000 like it is here 141 00:07:01,000 --> 00:07:02,333 or somewhat flat 142 00:07:02,333 --> 00:07:03,299 it's like a 143 00:07:03,366 --> 00:07:04,399 has a few pattern 144 00:07:04,400 --> 00:07:06,033 but we will get away from it 145 00:07:06,100 --> 00:07:06,800 get away with that 146 00:07:06,800 --> 00:07:07,866 it's just a wall 147 00:07:08,866 --> 00:07:11,599 but let's say you have a blue screen or a green screen 148 00:07:11,766 --> 00:07:15,766 it would be perfectly fine to use the original noise 149 00:07:15,766 --> 00:07:18,333 when you add it back on the background 150 00:07:18,333 --> 00:07:19,799 as long as it's even 151 00:07:20,400 --> 00:07:21,500 because you see 152 00:07:21,733 --> 00:07:23,399 if you had a painting here or something 153 00:07:23,400 --> 00:07:25,766 you can see you get a pattern in the noise 154 00:07:26,333 --> 00:07:27,899 from the shape you know 155 00:07:28,066 --> 00:07:31,699 so whatever you add back here in the background 156 00:07:31,700 --> 00:07:32,800 then if it's a pattern here 157 00:07:32,800 --> 00:07:34,866 you will get it on top of your background 158 00:07:35,666 --> 00:07:37,399 so it won't work 159 00:07:37,400 --> 00:07:38,866 then you will have to 160 00:07:39,266 --> 00:07:41,599 recreate the grain in some other way 161 00:07:41,600 --> 00:07:44,166 you know but for 162 00:07:44,966 --> 00:07:46,266 for this course we can 163 00:07:46,266 --> 00:07:48,299 we gonna add it back and keep it simple 164 00:07:48,300 --> 00:07:49,366 this is totally 165 00:07:49,366 --> 00:07:50,966 totally fine in this case 166 00:07:51,266 --> 00:07:55,566 so what we would do to add grain to our checkerboard 167 00:07:56,166 --> 00:07:57,499 we would just 168 00:07:58,733 --> 00:08:01,033 we isolated our grain right 169 00:08:03,400 --> 00:08:05,200 hook it up to the checkerboard 170 00:08:06,466 --> 00:08:09,466 just gain a bit you know 171 00:08:10,266 --> 00:08:11,599 and here you have it 172 00:08:11,766 --> 00:08:14,633 now we added our our grain 173 00:08:15,533 --> 00:08:18,999 so this is just to show you how to isolate 174 00:08:20,766 --> 00:08:24,499 the grain and we will further use this grain 175 00:08:24,566 --> 00:08:26,699 we will perform a retouch operations 176 00:08:26,700 --> 00:08:29,333 on the image that does not have the grain 177 00:08:29,333 --> 00:08:30,166 you know the noise 178 00:08:30,166 --> 00:08:33,066 we always want to do our operations without the noise 179 00:08:33,500 --> 00:08:35,266 and then when we finish our operations 180 00:08:35,266 --> 00:08:37,033 we gonna add back the noise 181 00:08:37,700 --> 00:08:40,066 and that way we keep the 182 00:08:40,266 --> 00:08:43,633 the original noise of the image or the 183 00:08:44,466 --> 00:08:46,633 the video or film that it was shot on 184 00:08:48,333 --> 00:08:51,966 so now that we finished doing our simple d 185 00:08:51,966 --> 00:08:53,866 noise in this clip 186 00:08:54,066 --> 00:08:54,966 in the next clip 187 00:08:54,966 --> 00:08:58,099 we're gonna continue doing tracking for her face 188 00:08:58,266 --> 00:08:59,599 I'll see you in the next clip 13333

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