【原创翻译教程】在Photoshop里合成vray输出的元素图 / 精品教程 - 中文汉化|专业汉化网站

离线 ati

楼主发表于： 2012-04-11 只看楼主倒序阅读

举报

0F

中国汉化网提醒您：

如转此帖，须经本站同意，另外请认真阅读《本站论坛管理条例》！

— 本帖被 lsray 执行加亮操作(2012-04-13) —

关键词： VRay 教程 Photoshop 技巧翻译

Compositing V-ray Render Layers inPhotoshop

在Photoshop里合成vray输出的元素图

作者：Ahmed Fathi

教程来源：http://cg.tutsplus.com/tutorials/chaos-group-vray/compositing-v-ray-render-layers-in-photoshop/
感谢人家写了这么好的教程

Twice a month we revisit some of ourreader’s favorite posts from throughout the history of Cgtuts+. Today we’retaking a look back at Ahmed Fathi’s tutorial on compositing Vray render layersin Photoshop. This tutorial was originally published back in July of 2010, andgenerated a lot of interest. So if you missed it the first time, why not checkit out now.
一直以来，Cgtuts+每月都会有两次回顾读者最感兴趣的帖子。今天我们来看一下Ahmed Fathi的vray 分层渲染并在ps中合成的教程。这个教程最早发布于公元2010年7月，并且引起了很多人的响应。所以，如果你错过了第一次，这次为什么来细细研读呢？

In this tutorial Ahmed Fathi takes a lookat how to composite together V-ray render layers using blending-modes and masksin Photoshop. Once completed, this process allows you to change or tweak anyaspect of your image in seconds without having to re-render a thing! Ahmed also covers a few extra post production techniques suchas Chromatic Aberration and Depth Of Field, as well as how to emulate aCross-processed look.
在这篇教程里演示如何将渲染的图层通过Photoshop的图层混合模式、蒙版来合成。一旦渲染完成，你就能分分钟改变或调出你想要的任意结果，而不用重新渲染！

Step 1
As this is a compositing tutorial, not alighting/rendering tutorial, I’ll assume that you have at least a basicknowledge of V-ray, and that you are able to render out your own scenesalready. We’re going jump straight ahead into setting up the different renderelements for the compositing process.
In order to make V-ray render out thedifferent layers, we first have to enable them in the V-ray Render Elements tabwithin the Render Settings window. Once in the tab, we want to enable thefollowing render elements as shown :

这是一个合成教程，而不是灯光或者渲染教程，我假设你有基本的vray知识，并且能渲出你的场景。我直接跳到如何设置vray元素以用于接下来的合成过程。
为了能使vray渲染出不同的元素图层，先得在vray渲染设置窗口的Render Elements选项卡启用各个元素，设置如下图
VrayDiffuseFilter漫反射滤光器
VrayMtlID材质ID
VrayObjectID对象ID
VrayRawGlobalIllumination Raw全局照明
VrayRawLighting Raw光照
VrayRawShadow Raw阴影
VrayReflection反射
VrayRefraction折射
VraySpecular高光
and VrayZDepth.景深

Most of these elements don’t need much work to getthem right, but we are going to need to take a few steps to set up theVrayMtlID, VrayObjectID and VrayZDepth layers.
其中的大部分vray 元素默认设置就能得到不错的元素图，但其中材质、对象ID、景深元素还是需要我们做些工作滴！

Step 2
We’ll start with the VRayZDepth element.The ZDepth layer is a black and white map that is used to tell Photoshop howfar each object in our scene sits away from the render camera – the further theobject is from the camera, the darker it will appear in the Zdepth layer.Typically the Min value is used to tell the compositor which objects will be infocus.
In order to correctly setup a Zdepth map,we have to adjust the Min & Max distances that V-ray should calculate, andtherefore what appears as white (the Min value) and black (the Max value)within our scene. Select the VRayZDepth item in the elements list. At thebottom you’ll see the zdepth min and zdepth max values we need to adjust.
我们从VRayZDepth 元素开始。ZDepth图层是用灰度表示场景中的物体离渲染摄像机的距离，颜色越白离相机越近。为了vray能正确计算景深图，需要测量你想表现景深效果的远近。从元素列表选择ZDepth，在底部有景深最小值和最大值参数，需要你调整。

Step 3
To get a sense of what values you shoulduse for the min and max values, you should use a tape helper object to measurethe distance between the camera and your closest and furthest objects.
To simplify things, let’s say sphere #2 in theimage below is the closest sphere we want to be in focus, whilst Sphere #6should be slightly out of focus. We would use two tapes to measure thosedistances and put them in as your Min & Max Values.
为了得出场景合适的景深最小值和最大值，你会用到helper里的tape对象来测量最近，最远，以及相机之间的距离。
简单起见，下图2号球我们想要它在焦里，并且是最近的（再近一点就会模糊的那种效果），同时6号球是最远的，有一点点出离焦（这一块照着英文意会吧，没译好）。然后用两个tape测量它们之间的距离，以作为min , max值

上不了图了。。照着原网页看图吧

Step 4
Here’’s what the VRayZdepth render for thiscomposite looked like after setting the Min to 20 and Max to 70 meters
本例中的值大概min:20 max:70米（是说房子的场景，不是上图的六个球）

Step 5
The VrayMtlID render layer creates an imagewith a different solid colour for each material in your scene, and to use it,we need to adjust your material IDs. In your material editor you will find anicon with the number 0 on it (as shown below.) If you click the 0, a grid ofnumbers from 0 to 15 will appear – this is your material ID. Go through yourscene and apply a different number to each material you want to have adifferent color in your MtlID element. As we have 16 numbers, we can have 16different materials appearing in our MtlID render layer.
材质ID图层是由场景中各种材质对应的实色组成，为了能适用，需要调整材质ID，材质调板上有个0的图标，按下后有个0-15的网格显示出来，这就是材质艾迪。检查场景中的材质，并赋予不同的材质ID，16个数字对应16个不同的材质。

Step 6
Here’’s how your VRayMtlID element would look likeafter setting up the ID’s for each required material. Your result may have different colors,but the important thing is that they are separate from one another.
最终分配了材质ID的元素图会像下图。你得到的成图也许颜色和本案中有所不同，但重要的是，不同材质的物体已彼此分开。

Step 7
VRayObjectID element is just like materialIDs element mentioned above, but it outputs a different colour based on thedifferent objects in your scene – the Materials are irrelevant to it. To setthis up, right click your desired object, select Object Properties and giveeach object a different number in the Object ID field in the G-Buffer section.
对象艾迪元素和上面的材质艾迪类似，对象艾迪的不同颜色是基于场景中不同对象输出的，材质都无关紧要。配置，右键对象——对象属性，在G-Buffer框的材质艾迪上给出不同的数字。

Step 8
I only needed to use the VRayObjectIDelement in order to make selecting the cars and the surrounding buildingseasier in post. Here’’s how it looks after rendering.
我用对象艾迪只是为了能更方便地选出车辆以及周围的建筑。下面是渲出的对象艾迪图

Step 9
The last element we need to add in is theambient occlusion layer, but as you’ll remember, this wasn’t added to ourrender elements list in V-ray. We actually need to render it separately afteryour initial render has been completed, but don’t worry, there is an easy, fastway to do it!
Apply a VrayLightMtl to your entire scene,and then add a VrayDirt map into the color channel. You’ll then need to go inand tweak the dirt map settings until you get a good, clean result result. Forthis scene, I got a nice looking AO map with a Radius of 2 in my dirt mapsettings.
Note: if you use a V-ray Physical Camera inyour scene, (which I highly recommend,) you will need to turn off your Exposureand Vignette options in your camera settings in order to render the AO passproperly.
最后添加的元素是环境阻塞层，（用AO插件吧，这不译了）

Step 10
After a couple of trials with the dirt mapradius, I ended up with the following AO element. I also increased the dirt mapsubdivs from 8 to about 64 which helped to smooth out the final render, and itstill rendered out relatively fast.
经过几次不同的污垢半径尝试，我完成了AO 元素，我也将细分从8增加到64来得到更平滑的的结果，AO出图还是很快的嘛

Step 11
Just a couple more steps and we’ll be readyto fire up Photoshop and start compositing! But first, how do we get all ofthese elements out of V-ray? For starters, there is the obvious way of savingthem to file one at a time, however there are infact two techniques used forsaving the different channels to disk and we’ll cover both of them in the nextfew steps.
The first method is really helpful,especially for those who use a Linear Workflow, and that is saving all therender elements out into a single .EXR file. To do that you should first selectEnable built-in Frame Bufferfrom the V-Ray::Frame buffer menu, and then turn onthe “Render to V-RayRaw image file option ”below. With that done, we can click browse.
仅仅几步，我们已经准备好启动PS开始合成啦！但是，怎么得到这些vray渲出来的图形文件呢？（接下来就是说通过几步设置，使渲染完成后的图像自动保存到预先设定的磁盘位置上）

Step 12
Browse to the place where you want to saveyour file, and enter your file name making sure to add the .EXR extension ontothe end of it. Then select All Files(*.*) as your file type. This will allowyou to save your render as a single .EXR file that contains all your differentrender layers.
If, however, you try to open your new .EXRfile in Photoshop, you will only see the first render layer. To see the others,you’ll need to install a commercial Photoshop plug-in called ProEXR, and thatis why I prefer the next method!
Note: EXR files burn in any gammacorrection as it assumes you’ll be using a linear workflow. If your imageappears washed out in Photoshop, go to Image > Adjustments > Exposure,and set your Gamma to 0.454, which is the inverse of 2.2 (calculated bydividing 1.0 by 2.2).
（这是作者介绍的一个方法，保存成.exr文件，PS读取.exr文件还需要一个ProEXR的插件，貌似收费，作者下文还介绍了一个他首选的方法——Split render channels。）

Step 13
The second method (my preferred one) is tosave all your elements out manually as .TGA files. I prefer TGAs to JPGs asthey are 32-Bit, and can hold much more color Information than the normal 8-BitJPG file. Another bonus is that they support having a built-in alpha channe.
Note: if you have never used the V-RayFrame Buffer to render out elements before, you can find all of your elementsin the top left drop down menu that says Diffuse by default.
第二种方法（我的首选）是不用再处理的保存方法，像tga文件，jpg和tga我更喜欢tga的原因时32位的tga能保存更多8位jpg不能保存的信息，另一个原因是tga能保存alpha通道。

Note:

Step 14
Now that you have finished our 3Ds Max andV-Ray Part, fire up Photoshop and let’s start playing! First a small tip to help you get all of thedifferent files into layers in one Photoshop document. It is a script builtinto Photoshop that will stack your files as layers and arrange them inalphabetical order. Go to File > Scripts > Load Scripts as Stack, andthen browse for your files, select them all and click Ok. Some people preferbringing in the elements one at a time, however I do prefer this method myself.
既然咱已经完成了3ds max和vray部分，那就启动Photoshop，一起摇滚吧！
先说一个小窍门帮助你将所有不同的文件用Photoshop打开成一个文件。这是Photoshop的一个自带脚本，将所选的文件以图层的形式并按名称升序排列。
步骤：文件>脚本>将文件载入堆栈，然后浏览你需要载入的文件，选择并点击确定。当然，一些人喜欢逐个载入这些文件，但我还是喜欢我的这方法，嘿嘿

Step 15
This is actually your first step incompositing! We’ll start by turning off the visibility on all layers except forthe Z-Depth. Select all of the contents of this layer (Ctrl+A), copy them(Ctrl+C) and then switch to your Channels tab.
Some people prefer adding a little bit ofGaussian blur to their Z-Depth before using it, however as most of my clientshate DOF, thinking that it’s a needless loss of detail, I do not do this! Inthe end it is a matter of personal preference.
这才是合成之路的第一步！首先关闭除ZDepth以外的所有层。Ctrl+a选择这层的所有内容并复制，然后粘贴到通道去。
一些人的做法是给ZDepth通道图一点高斯然后再用它，然而我的大部分客户不喜欢景深，认为那会造成细节的不必要损失，我不能这么做！最后，它（DOF）是种个人爱好。

Step 16
Once in the Channels tab, first check tosee that your original Alpha layer is present. If it isn’t, open up theDiff.tga file seperately, switch to it’s Channels tab, and drag and drop theAlpha channel from there onto our Z-depth image – an Alpha1 channel shouldappear. You can then close the Diff.tga file.
Now, click the Create New Channel and pasteyour Zdepth render into it. You can then switch back to the Layers tab anddelete your Z-Depth layer as we don’t need it anymore.
打开通道面板，首先看到的是你的原始Alpha层，如果没有，再另外打开漫反射元素图，切换到通道面板，拖动Alpha层到ZDepth图上，Alpha1通道就成了。然后就可以关闭漫反射元素图了。
然后，新建一个通道图层，并粘贴ZDepth图层。然后你可以回到图层面板删掉ZDepth图层，因为俺们不再需要它

Step 17
To follow along with me in the next fewsteps, first arrange your layers in the order shown. The RGB layer at the topof the stack is the raw render, straight out of V-ray. I’ve kept it there sothat we have something to compare to at the end.
接下来将图层按序排列在面板上，RGB层放在最顶层，是又vray直接渲染输出的。用作对照。

Step 18
First, duplicate your Diffuse layer byright clicking it and choosing the Duplicate option, and then move the copyunder your RawLight layer as shown.
首先复制Diffuse（漫反射）层并拖至Rawlight层下

Step 19
Turn on visibility for your Diffuse layerand your RawGI layer. Set your Raw GI blending mode to Multiply.
使漫反射层和RawGI层可见，GI层模式设成正片叠底。

Step 20
Next hold down the Alt key and clickbetween you RawGI layer and your Diffuse layer (the cursor will change intothis two circles and an arrow icon when you’re in the correct place). Thislittle bent arrow that appears indicates that this layer (RawGI) is onlyaffecting the Diffuse layer. This is called a clipping mask, and you can findout more about it using a simple online search. There are tons of tutorials outthere!
剪切蒙版，diffuse作为蒙版。RawGI只对Diffuse图层起作用

Step 21
Using the same technique that you learnedin the last two steps. Enable visibility for both “Diffuse Copy” and RawLight. Set the blending mode for theRawLight Layer to Multiply then use it to only affect “Diffuse Copy” layer (Alt+click between the two layers).
依葫芦画瓢，应用到RawLight和Diffuse copy图层上，RawLight层混合模式为正片叠底

Step 22
For the sake of better organization, groupboth the Diffuse & RawGI layers into a group, and the Diffuse Copy &RawLight layers into a different group. To create a group you can click on thesmall folder icon at the bottom right of your layers (I should say sorry forpeople who actually know all this basic stuff. I am just trying not to letanything pass by for beginners too!)
为方便管理，用文件夹吧，看下图，不译也懂的

Step 23
Change the blending mode the of “Diffuse + Raw Light” Group to Linear dodge(Add). This adds the information contained in this group to the informationcontained in the group “below”. The image should start looking more natural now.
更改Diffuse + RawLight组的混合模式为线性减淡（添加）。$%^%$#$%&**^%。图看起来开始逼真起来了。

Step 24
At this point things look much better, butyou may be asking yourself – “why is my glass black?” Well, most of the glass’ information in the final render is contained in the reflection andrefraction passes, and we haven’t composed them just yet.
So, enable visibility for your ReflectionLayer and set it’s blending mode to Linear Dodge (Add) and see the difference.One of the main benefits of having a composite like this is that, for example,on this reflection layer you can paint/paste in any reflection you want toappear in the windows.
Note: I should explain a little bit whatlinear Dodge (Add) blending mode does. It adds the color information of thepixels to each other. We know that for example Pure White is 255 and Pure Blackis 0. So adding pure black adds 0 whilst adding pure white adds 255. Therefore,pure black (0) + pure white (255) = Pure white (255), Mid grey (125) + Mid grey(125) = White (250) and so on. That is why when you use this mode you no longersee any of your black, because it’s a zero-value color.
现在看起来要比先前好些了，但是你可能会问：为什么我的玻璃是黑的？其实，大部分的玻璃信息在最终渲染后都被包含在反射折射信息里了，只是还没有混合它们而已。
So，出来吧，Reflection Layer，混合模式设成线性减淡（添加），看出差别了吗。最主要的好处是合成like this is that，举个例子，在这个反射层，你能尽情地绘制，粘贴你需要的窗户反射信息。
Note:我应该解释下神马是线性减淡（添加）混合模式。它增加了像素之间的颜色信息，我们知道纯白是255，纯黑是0,纯黑和纯白以这模式混合后还是纯白，中度灰（125）+中度灰（125）=白（250），这就是为什么当你用这个模式混合时不再看到任何纯黑，因为它（纯黑）是Zero-value color（0值色）

Step 25
To illustrate how easy it is to edit any ofyour elements (color correct it, adjust exposure, etc…) I will assume that I now want to change thereddish color of my stone texture. If we hadn’t used this multi-pass compositing method, wewould have two options:
Adjust the texture itself in Photoshop andthen re-render, perhaps completing a dozen test renders before you weresatisfied with the result, and could render a HQ render. This obviously takes alot of time and patience!
Using selection tools in Photoshop, wecould select the texture and apply the desired corrections to it. The firstproblem with this is that selection is really a very tedious job, and when youcome to add color corrections, you’’ll be affecting all of the image – your shadows and highlights will look oddand overall things won’’t look that good.
为了说明使用元素可以多么方便地编辑你得任何效果（颜色改正，曝光调整等等等等），我将假设我想把石材的质感调得红一些。如果不使用多通道混合的办法，我们有两个办法解决它：
1．用Photoshop调整材质贴图，然后再渲染，
2．Photoshop里作选区

Step 26
Well, remember that element called MtlIDthat we rendered out? It is time to enable that layer. The only use for thislayer is to create MUCH faster selections with only a couple of mouse clicks.Using this layer we will easily select our stone texture in no time at all.
亲，还记得大明湖畔的MtlID吗，是时候召见她了。她的唯一用处就是轻点几下鼠标超级快速做出选区。

Step 27
With the layer selected, go to Select >Color Range. Then use the eye dropper tool to sample the magenta color on ourmaterials layer that represents the ID of the stone texture. With that done,press OK and you have your selection.
Go to 选择>颜色范围。用取色滴灌工具选石材对应材质通道图里的颜色，点击ok，就ok了。

Step 28
Now turn off the visibility of your MtlIDlayer. Then while keeping that selection active, add an Exposure adjustmentlayer. This will create the adjustment layer and automatically add a mask sothat it only affects your Stone. It doesn’t matter where you have this layer now, wewill move it later on.
关闭材质通道图，保持选区的有效性，添加一个曝光调整图层，这会自动创建一个调整图层和一个匹配选区的遮罩。图层现在的位置不要紧，等会会移动它的。

Step 29
In the Exposure adjustment layer’ssettings, set your Exposure to +0.80 to brighten it up a bit.
到曝光层，把曝光调到[font=&amp]+0.80 来提亮一点

Step 30
As we want this layer to affect our diffuseonly, you will have to place it right above the Diffuse layer (remember thatone inside the group?) Move it into position as shown and remember to Alt-clickbetween the layers to make it only affect your Diffuse channels.
Just as a side note; whether you add it toyour Diffuse layer or your “Diffuse Copy” Layer shouldn’t make a difference at all. Some corrections might require you toplace them on both Diffuse layers though, and it’s just a matter of trial and errors untiltill you fully understand it all.
剪贴蒙版还记得吗，如果只想影响漫反射层你就会用到这个了。

Step 31
Take your Refraction layer one step down sothat it sits right above your Reflection layer. Enable both and set theirblending modes to Linear Dodge (Add). I’ve added another Exposure adjustment,with the Exposure value set to +0.80, to my Refraction layer as I wanted it toappear a little brighter.
Note: Editing the Refraction layer is oneof the best ways to adjust the tint of clear glass in interior renders.
可见折射层并移到反射层的上一层，混合模式同样是线性减淡（添加），我还给了个曝光度调整图层，曝光度+0.80好让折射看起来亮点。

Step 32
Enable your Specular layer and once againchange its blending mode to Linear Dodge (Add).[size=font-size:10.5pt,10.5pt]使高光层可见并更改混合模式为线性减淡（添加）

Step 33
Duplicate your Specular layer (Ctrl+J) –notice it still has Linear dodge (Add) as its blending mode – and then with thecopy selected, goto Filter > Blur > Gaussian blur.
Ctrl+J复制高光图层，注意复制的图层连同混合模式一起复制下来了，它还是线性减淡（添加）模式，然后滤镜>模糊>高斯模糊。

Step 34
Set your Gaussian Blur Radius to around 2.5pixels and press OK. Set your layer Opacity to around 65%. This should helpcreate a specular bloom effect around your specular highlights.
Note: These values are not constant numbersas they depend on your resolution, and of course your taste. Just don’t over-do it!
本案中的模糊参数是2.5pixels（非定值，因图而异），点击确定。调整图层透明度65%左右，这会产生一圈不错的高光光晕效果。
Note:见上文括号内容

Step 35
Enable your RawShadow Layer and invert it(Ctrl + I).
点亮Raw阴影层，Ctrl+I反相

Step 36
You might be surprised that your shadowsare bluish but this is in fact normal, they always are during the day becauseV-ray’s GI has a bright blue sky. Set your RawShadow layer’s blending mode toMultiply and it’s Opacity to around 20%.
你也许会疑问这阴影怎么是蓝色的，但实际上这是正确的，因为vray中室外产生GI的是蓝天。更改图层混合模式为正片叠底，不透明度20%左右。

Step 37
This step is not a must but I likeenhancing my shadows a bit as it gives the image a bit more contrast. Add a “Color Balance” adjustment layer,making it only affect the RawShadow layer, and with the Tones value set toShadows, give your shadows a reddish tint. Set the opacity for this adjustmentlayer down to around 20%, although again, this is just a matter of personalpreference.
In other scenes when I find my shadows alittle too sharp for my taste, I add a small radius Gaussian blur filter to theshadow layer, which helps smooth out those hard edged shadows a lot.
这一步不是必须做的，但是我想加重一点阴影来让图像对比强烈点。添加一个色彩平衡调整图层，让它只影响你的Raw阴影层。调整阴影的色调，让它偏红点，透明调整成20%左右。再次提醒，这只是个人偏好
我在其他的一些场景测试中感觉阴影过于锐利了，就给点小半径的高斯模糊，这能让你的硬边阴影看起来平滑一点。

Step 38
Now enable your Ambient Occlusion layer.Set its blending mode to Multiply and its Opacity to around 10%. AmbientOcclusion shouldn’’t be too obvious in your final render, as it just helps you enhancethe look of the little details in your final image.
再来点亮你的AO层，图层模式正片叠底，透明度10%左右。环境光阻挡的效果并不要很强烈，我们只是需要它来给图像增加一点细节而已。

Step 39
At this point, we are pretty much done withthe compositing! To sum it all up here’’s a simple equation that I used until I memorised the differentblending modes. If you follow it through, you’ll see it exactly matches whatwe’ve done in this tutorial!
至此，咱差不多快完成合成了。下面有一个简单的公式用来在你完全掌握混合模式前辅助。挨个看完你会发现这和咱的教程是多么的匹配。

整完了，余下的就是一些后期教程，看图就应该明白了
译得不好，就一个大概意思
有问题请留言，大家共同学习探讨

共1条评分，金币+10

lsray

金币 +10

很好的后期处理教程，希望楼主尽快完善

2012-04-12

分享到