Archive for the ‘(*postProcessing tutorials!’ Category
Our new (*chanimations VintageCircus release ‘A Tale of Fortune’ comes again with a neat little phototrick – and here’s how you make good use of it when working in a graphics program on the picture! Click the image for a full screen view!
Some of the steps are not described as detailed as they could be, I am assuming you are a little bit familiar already with the filters in your graphics programm of choice ;-) I am not quite as familiar with the exact names and where you find filters in all the programs, but chances are if you search the web you will find out fast if your program can do a feature or not – or what the workaround might be!
If you have any questions regarding a step, feel free to leave a comment here and I’ll try to clarify – if I notice a question comes up repeatedly, I might edit this post in the future to add the FAQs ;-)
Have fun recreating!
I’ve ran across a few different blogposts lately that took on the issue of ‘good picture taking’ in SL and what it means for the respective writers, as well as some nice posts picking up the challenge trying to explain ‘their way’ of snapshooting. So to not reinvent the wheel, have a look at Chic at Phil’s Place “The Photo Dilema” for a rather nice analysis regarding the use of shaders and resolution, or have a look at Luna Jubilee’s “How to take a High Res Snapshot” tutorial if the term ‘high res’ doesn’t sound familiar to you at all. And last, if you’re more the definition type, a nice write-up in Vint Falken’s “SL Photography: Tips and Tricks” which explains you exactly what each SL menu option does.
I liked the general idea behind all these post, so I decided to write another picture tutorial post to share a few secrets of my own when it comes to ‘raw SL snapshots’ using the means provided by the client. I am a more visual person, so I went and took a bunch of pictures for you, and you’ll find some explainations along with them for the whats and whys. There is no rights or wrongs in my opinion, just a very broad spectrum of possibilities to match your ideas and achieve a desired purpose. And for me it usually works best to just see some examples! Click the images for a larger resolution so you can read the texts ;-)
Choice of Angle and Zoom Tools inside Second Life
Probably the first step to overcome, and it comes with a zillion of different choices! There’s a few things to consider when picking your angle, but the most obvious is called ‘purpose’ – what are you trying to highlight in the image. Is it a landscape portrait to highlight a scenery or build? Is it telling a story between two characters? Is it artistic expression with dynamic angles? Or is it highlighting a person and their traits? Or a fashion portrait in which you like to show details of an SL creator’s work?
True enough, you’ll get a lot of cool pictures with just starting to snap away. But if you put some thought into what you like to achieve beforehand, you will need way less snapshots saved to your hard drive, trust me ;-D
So in the following a bunch of different angles and tries. I wanted a squared format, so that was part of my considerations from the start.
Your best friends are the zoom tools: That’s CTRL-0 (zero) to zoom into a scene, CTRL-8 to zoom out of a scene and create a wideangle format appearance – or something completely wild if you zoom out too far ;-D And if you need to go back to default camera position, CTRL-9 will get you there!
Why zooming? It allows you to balance the sizes of things, without having to move them around. I picked an avatar in front of a very detailed background to demonstrate what I mean. If you compare the sizes of the avatar’s head or length of arms, etc., with the size/position of the items in the background, you will notice that depending on zoom in or zoom out, the ratio changes and the avatar appears smaller or larger. As well, the angle in which you look at things changes depending on how extreme you zoom.
I am using angle and zoom as well to cover things I don’t want in the picture with my avatar. So you will notice for example a bright orange poster in the background of some examples, while others cover it up without having to move it out of the picture.
Shaders and Graphic Settings
The higher you can go, the better. But that doesn’t mean you need to switch to ultra and get a shadow compatible viewer along with a new computer to get some nice results. True enough, Ultra with shadow settings enabled are gorgeous if your computer can push it, but not all PCs/graphic cards survive to even launch the client. And since mine is one of those computers who translates ‘shadow viewer’ with ‘blue screen’ I’ll skip those in here and keep it more basic.
Bump and Shiney is the first ‘quality step’ to enable, and while it does not allow you to see any glow or light sources yet, it at least adds a soft metallic shine to prims that have bump mapping enabled. (See the collar and wheels of the train)
Basic Shaders should be possible to run even with an older computer, and they give already very decent results. Glow appears now, light sources are rendered (and with them you can influence your scene as well if the sun alone doesn’t get it for you). Moving the position of the sun in SL’s sky settings is pretty much ALWAYS a good idea. Write it big on your computer screen if you tend to forget it. It’s like your portable photo studio to make sure light falls on those parts that need it.
Atmospheric Shaders – the Windlight. Safe to say, likely the most revolutionary improvement of the SL client when it comes to picture taking. If you can run it, run it. If you can’t run it, consider to at least turn it on for a short moment to take a pic or two when the outcome of the pic really matters to you. ;-D
Expressing and enhancing…
If your computer doesn’t like atmospheric shaders, but you like to use them, only turn them on when your angle is set and all textures/prims are rezzed and loaded. If you have a lightsetting you like, save it. Being able to quickly get to it will help you a lot, and over time you will have a small selection of your favorite lights available so the more you use it, the more time-efficient you will get with it. (I used to leave the graphics performance tab in preferences open and minimize the window on my old computer, so I had quick access to turn it on and back off after the pic was taken)
Again, purpose is our main theme when picking a lightsetting.
If you like to go for a fresh beauty photography feel, don’t use extremely harsh light settings which make the flaws of SL’s avatar mesh screamingly obvious (unless you want to spend the time in a graphic program to fix it). On the flip side, if you want to go with drama and dynamic, try to give it that quality with contrasted shadows and light siluettes on the avatar by moving sun position coming from left or right instead of a direct frontal sun position.
Another aspect for light choices are the colors in your picture. A dark skintone will cope way better with harsh lights than a very pale one does. Like in my example, some pictures are over-exposed and the definition of face and texture of the white dress get lost. So if your main interest is the skin and dress, likely not go with that one ;-D
You will also notice that sun position ‘shapes’ your avatar, most obvious part is the face. In the example on the right side, top row, the sun shines from a high noon position down on the avatar – the shadows change the overal look dramatically – not quite what I was going for here either, although I like the intense fog in the background a lot!
In the end, I went with more subtle lights, choosing a little foggy feel for the background to draw the focus of the viewer towards the avatar.
Beyond the in-world light settings, there wasn’t much needed to be done in a graphics program. The only thing I did was enhanced the contrasts slightly, add some very minor blur, and added a frame to it.
And voila… This is what I ended up with!
Outfit: !dM deviousMind “Little Lenore”
Shoes: Stiletto Moody “Bare Judy” in white
Eyes: !dM deviousEyes “AngelDust” **ICEBLUE**
This time, I am focusing on alpha-glitches in avatar bodies and surroundings.
In most cases we realize their existance only then when out of all things, our favorite prim-avatar, demon-leg, pair of boots/shoes, or whatever else it may be is the cause for them. They occur, when alpha meets another alpha (or semi-transparent texture), for example with the water we’re standing in, “reflective” floors or floor shadows, tree leaves, groundcovers, or any other sort of semi-transparent texture use. Sure the list goes on, and some of it has most likely run across your ‘camera’ already.
We’ve grown pretty much accustomed to invisible prims ever since we realized that nobody wants to walk around in noob-feet. And the higher, more extreme, more sexy, the better (hey I might speak for myself here, I realize this! *laughs*). What matters more though: you don’t ever need to take these shoes off if they react in some manner with the respective surrounding. – And I’d rather taken no picture than take them off!
There are some things that you can pre-fix already when taking the picture simply by knowing how…
I’ll show you some of these tricks here that I used for a quick-fix in a picture shooting the other day!
This is what we started out with, I picked my personal weak spot, high-heels (and super-gorgeous ones at that :-D) and combined it with an interesting angle that made sure I have what I’d like to show you clearly visible to demonstrate how to prepare the fix:(I watermarked the images this time, and since it was just taken on plain gray added a bit of a vignette to it – it looked all soooooooo boring uni-color without! As with the last online tutorial, click on the image to open it in a separate window; it will also allow you to navigate forward and back when clicking the arrow that appears in both upper left and right corners.)
In the future of SL the famous ‘transparency blanks’ that occur when alpha-prims meet other alpha-textures on the ground, trees or avatars will hopefully be solved by the new alpha layers that were introduced this week with the viewer 2.0 beta. As I don’t run this viewer on my computer and since likely not all old shoes will be ‘out of style’ instantly or updated right away with the viewer changes, here’s a little tutorial on how you can prepare your pictures already during the shooting inside SL (without the new alpha-layers) to make the fix in a graphic program afterwards as simple as 4 minutes work – and with hardly any risk to mess it up at that! :-D
As usually, there are many ‘right’ ways to get to the goal, so the technique I’ll describe in the following doesn’t apply to all problems. But it is a great help when you’re in a studio surrounding, or if you have to use a model as the camera rotation point to get the right angle you want, and with that having one of the models ‘get out of the picture’ physically is a no-go to maintain the camera perspective (as it was the case for me here.)
So next thing we need is ‘Advanced’ menu
– if it doesn’t show yet in your top browser bar, toggle it on/off with [CTRL]+[SHIFT]+[ALT]+D. Then Go to Rendering, and mouse over to select “Hide Selected”. It shows a little x when it is activated.What we will do is simply ‘hide’ everything that appears in front of the kneeling girl so we have a second picture to lie under our first one without any holes. When editing the shoe now (applies if you are the owner of the attachment) it turns invisible, only the highlighting lines show still, and edit window opens. This works slightly different when you are the person not wearing the item, then you need to do right-click > more > more > inspect to select the item.
We still have the foot shape of the SL avatar now – if you are not owner of the shoe attachments, you can just go on and ‘select’ the avatar too so it goes invisible. Since this was not an option for me in this case though, and I am using a trick in Emerald Viewer, “Render Avatar Invisible” which can be found in Preferences > Emerald > Avatar. (Actually I am sure this option exists in the regular SL viewer too somewhere cause I think I used it in the past, but I don’t know where it is hidden so you’d need to confront the wiki on that question!)
Anyhow, my avatar is gone now too, and while I took the snapshots here specifically WITH the UI and stuff to show you what I did, the second picture I saved to hard drive (the first was the image we started with) looked like this – little bit creepy, but efficient:
In my graphic program, I moved both images into separate layers of the same file – ‘Background’ is my second inworld shot with the partially hidden avatar, and I just pulled the other one with holes in the avatar on top into a separate layer (see layerorder screenshot below). It is important that the two pictures align with one another, and to have the exact same angle and no moved camera in both pictures! (also in case you never needed to use it, most graphic programs have an alignment tool that can be applied to multiple layers to make sure borders cover one another exactly)
And the final magic is as simple as that: use your magic wand tool, select the grey area around the shoe on the top layers, and delete them out of the layer.
You have created a hole now, that allows to see the layer underneath :-)
No need to know how to draw, and with a little bit of preparation inworld already all is fixed in a few minor post-processing steps. And from here on, you can go on with your actual image treatment to make it sparkle! …which I did, as you saw!!!
Chair with poses for both models:
“ShoeFetish” (FetishCabaret SET 23) by (*chanimations
“BloodRayne Costume” by *Deviance*,
BARE Robin Ankle Boots by Stiletto Moody
Latex-Lingerie by *HD* Hugo Design,
Ballet Boots by *Dilly Dolls*
Glamorful and shiney and…. post-processed.
There’s a lot to be said about picture treatment in a graphic programm. And there’s many out there, from the well known professional (and rather pricey) ones, down to the free shareware. Whichever one you pick though, basic steps of treatment are always the same, especially when we’re looking at something as plain and rough-edged as a SL screenshot – I’ll try to lead you through some of the essential steps how you can enhance your images, and will try to stay as basic as possible with my explainations so that they apply to more than just one programm. Please take note, you will need some basic knowledge of graphic programs to repeat the steps, the exact ‘how to do’ is something I can’t teach you that quickly. With this post, I’d like to make you aware of the overall concepts, and lead your eye to the areas that need be worked on, and show you steps that help achieve this. (And you can use that as a basis to do more own research)
One of the most important things you need to realize right from the start is not so much that an expensive programm is neccessarily ‘better’ or free is neccessarily ‘crap’ –
Your eye for the overal picture you’re creating is what matters most in the process, not the programm you are using as the tool.
If you want to try with freeware first, I’d highly recommend you the GIMP. It’s shareware, has nearly the same features as photoshop, and you can google a damn lot of information and help in the net if you get stuck somewhere as there is a lot of GIMP community forums.
So that being said, let’s take a look at the processing!
This is the picture I am going to start out with, it’s a screenshot using windlight settings and I like working with two lightsources (which is why you see light both on the left AND right side of the thigh for example. This technique comes from RL photography in a studio, where you usually work with at least two sources with different intensity.
Looking at the picture above you will notice a few things:
The screenshot looks flat. As well, the image is far too yellow, and the overall color balance is not good. When looking close at the avatar, the familiar ‘SL spikes’ of the mesh show where shadows/lights fall on the surface, as for example in the hand/wrist area of the streched arm. And finally, the familiar ‘straight lines’ where you would expect curves in a real human body – you see those best at the left uppper calf in the back.
So here is where the most work part beginns: masking the avatar out of the background.
Unless you are somewhat professional with this process, it can take a very, very long time. Which is why it is in most cases easiest to shoot the image in front of a greenscreen, bluescreen, or whichever other bright color that does not appear in the picture. If you did that, it’s as easy as a click with the magic wand and some (usually minor) adjustments. Use of SL depth mode comes in handy here too, it can shorten the process tremendously! (please see my “DurgaDevi” tutorial where I used depth-mode for masking)
If you didn’t do any of the above, it’s gonna require handwork, and a mix of different tools will likely be the best way to go about this. If you have never worked with masks before, this would be the time to familiarize yourself with them, as they make things ridiculously easier. Either copy the av on it’s new layer, or most definitely SAVE the mask you created (To explain how to do these things would burst the amount of steps I’d like to cover here though, so please google away if you’d like the info!)
And in the following two steps you’ll see clearly why we wanted avatar separated from the background:
Take note how the background is blurred now, the front (flying hearts, avatar and pralines’ box) still stays sharp. This creates a more 3D-feel, imitating the depth of field your eye experiences. The eye can only focus on ONE point, everything else that appears far behind or far upfront of the focal point turns blurry – so that is what we want here too!
As well I adjusted color balance, contrast, brightness and satuation. All these things were done in multiple steps, best you apply each to an individual layer so you can undo things, or lower the opacity if one aspect is too strong. The general order: adjust levels (try auto-adjust, in many cases that’s a first great step), work with color-balance (to get rid of too much yellow, too much magenta, too much green), if required change the contrast, and if required adjust satuation. And if any of these steps leads to unwanted results, like skin turns too bright when you adjust the background, or colors turn over-satuated: Use the mask. That’s why we created it :-D
And we’ll also use this mask to create a shadow – this shadow basically is on a new layer UNDER the cut-out avatar layer. I filled the original mask with black, enlarged it slightly and moved it a bit to the right. Then I moved it underneat the ‘cut out’ avatar layer, set it to only 20% opacity, added blur, and there’s our shadow falling on the ‘wall’ and not onto the avatar :-D.
Looks much better already, doesn’t it.
If you flip back and forth with the picture before, you will as well notice a change in the SHAPE of the avatar. (Click the image and it will open in a larger window – click with your mouse in the upper left/upper right corner of this window for ‘previous’ and ‘next’ image)
Take especially note how the breast are lowered and made wider now – since SL avatar doesn’t follow gravity when lying, this was urgently needed to not make them poke out into high air anymore, but give more of a natural lying feel. As well, the hips are much wider, softer and more curved now. These things were fairly easily done by use of the ‘Liquify’ (Photoshop) or ‘IWarp’ (GIMP) tool.
The avatar turned more ‘curvy’ and most of the ugly spikes are gone. Take a look at Vint Falken’s video tutorial if you’ve never used either of these tools!
One more detail I added: a noticably whiter center of the starburst – and again my before created mask was the main thing that made this super-easy to add (same as the shadow before). With adding the lighter center and the darker corners (the ‘vignette effect’ of a camera), I as well adjusted the brightness/contrast again to create more dynamic in the image and balance it more out.
With that, we’re pretty much done, and the difference to the next image is only subtle. If you flip back/forth again, you will notice that I added heavier blur towards the outer edges of the image (again to create focus on the center), a black frame, and finally some noise which is imitating lens-blur: the background is heavily noisy, the focus in the front is smaller noise. You might know this behavior from RL photography maybe…
And here’s the end-result – YAY :-D
Shown in the image is my own avatar, posed with (*chanimations fruitFetish! #16 set “SweetValentine”, which includes the red ribbon wrapping, the chocolate smeer, a huge praline-box, poses and more stuff that’s not shown here (see linked post for full content listing or xStreet description). Only thing added are the super-gorgeous and matching-for-the-occasion high heels by Stilletto Moody (“Elizabeth”) – and I was readily dressed :-D
I made the picture originally cause I wanted to make a PRETTY valentine’s picture for Lucca – in the end added some more elements, lens flares and text/logo, and ended up with another little storeposter – which is the one including the text that you’ve seen first in this post.
So… hope this gave you some ideas and insight, and have fun applying some of these concepts to your own work! There’s a million different ways how to work on pictures, these are some of the core steps I do with mine. :-)
Also keep in mind, internet search sites are your friend and will bring you TONS of helpful forum posts and other things to help you find solutions to your graphic-problems!
And I’ll go take care of my Valentine now :-) *waves and poofs*
As so often before mentioned in the individual set introductions – there is a calenderPack! So, of course here’s as well a calender tutorial!
This is the final image we’re going to create (or well, my version of this final image. )
Of course in your case, it will look similar to whatever you started out with of course ;-) – So, the first thing to do is snap away! Take note, plan for some ‘calm space’ in the image where you’ll add your calender pages in the post-processing! Keeping this in mind from the start will make it a bit easier to do in the post!
Please click the following image to enlarge it for better reading:
This little post-processing tutorial will answer one essential question: How to turn 2 arms into 8! This is the final image we’re going to create:
And here’s how to do it! Please take note, in the following I’ll use the method of taking pictures in ‘depth mode’ – you can as well take the images in front of a greenscreen or bluescreen as we did in the CabaretBeauty tutorial. Using either of both methods has advantages and disadvantages, always keep that in mind!
I’ll start with advantages for each:
GreenScreen is something that’s portable and you’re not dependent on rezzing anywhere. As well, you have all freedom for a final composition of the picture, as you pick the background yourself.
Depth Mode allows you to use it in any situation (unless your computer crashes on it) – and when it comes to the overlaying work as will be described in the follow, it is extremely forgiving when it comes to edges, as there won’t be any fuzzyness and you don’t need to be as precise with setting things free as you need with greenScreen background – the background colors are all the same on all layers anyhow ;-)
But there’s also Disadvantages for each:
Depth Mode does not display ANY prims that are partially transparent – be it texturewise or prim wise which makes it extremely difficult to catch exact lines of hair for example! (this will not have any bearing on our work with DurgaDevi, luckily, which is why all attachments in this set are full prims so they show in the depth shot ;-) – and for partially missing hair in a cut-out, we’ve covered that with the ‘advantage’ section of depth mode already: all layers are the same anyhow, and the lowest layer has no cropping! )
GreenScreen is a royal pain in the a** to get out of half-transparencies. (like hair fringes!) – if you want to work with a solid background color to use the magic wand, you should pick a color that is close to your final background image that you will put underneath our 8-armed Durga. For example, if you want to put her in a forest, take greenScreen. If you plan to put blue clouded sky or water into the background, rather use a blueScreen. By doing that, the ‘un-clean’ parts in semi-transparent areas of your SL shot won’t be as bright obvious.
And well, here you go!
(please click image for larger view)
Above is the final image which we’d like to create – and in the following comes how you do it!
Just first off a few added extra tricks to keep in mind for taking your SL shots. As a rule of thumb, the better you know what you’re doing in advance, the quicker it will look great afterwards!
1. Know what you’re doing
If you go into your bathroom, bedroom, or any other mirror you have at home (even the surface of your TV will do for the basic understanding) – check what you actually SEE in the mirror depending on which angle you look at it.
if you’re straight up in front of it, on the same level looking inside: you see yourself straight upfront from the same hight level in the mirror as well.
If you’re sitting on the ground looking up into it… you see the higher ends of the walls if not even the ceiling. (imagine how your view through the glass would be from the backside: you’d be sitting on the ground as well looking up at that ceiling, right?)
If you make yourself aware of what angle will be seen, your images will turn out MUCH more real appearing than if you just take a random angle!
2. It’s a mirror – flip your reflection image horizontally!
Even if it would look okay otherwise – everybody knows a mirror in real life – any will notice “something is wrong” if you forget the flipping.
3. Blur/Desaturate the reflection a little
True, in RL it has the same colors – but with the 3D the eye can see, it is easy to understand that it’s a mirror. In a flat image, you need to be a little more helpful to the watching eye to show ‘distance’ to it. gentle blurring and desaturating is one way to do it.
Please click the following image to see large to better read the text:
I’ll search you some wiki links on how reflections work and post them here once I found them ;-)