Can someone give me tips for putting images together...?
Upirium
Posts: 705
i don't know how I did it wrong the last time, but I did, and I'm wondering how I can merge an image...
I have a large image I want to render but if I zoom out all the way I am going to lose a lot of detail. So I want to do two images and put the mtogether, one ontop of the other, but of course they need to be fluid.
Somehow I tried this once before and messed up. Things got uneven even though the only movement I did with the camera was up and down, not sideways...
So what do I need to do to put two images together to make them even...?
Daz 3D is part of 
Connect
DAZ Productions, Inc.
7533 S Center View Ct #4664
West Jordan, UT 84084
Licensing Agreement | Terms of Service | Privacy Policy | EULA
© 2024 Daz Productions Inc. All Rights Reserved.
Comments
What program are you using to render the images? Also, what program are you using to merge the images on top of another? The same program? Photoshop or Gimp?
When you say put one on top of the other, are you talking layering one over the other, or using 2 smaller images to make 1 bigger one?
In the case of the latter, if you have access to Photoshop you can do it by going to File > Automate > Photomerge.
In the case of the latter, if you have access to Photoshop you can do it by going to File > Automate > Photomerge.
Two images to make one bigger one. Like, the art I am doing right now involves wings, and I want to do the entire wings, but the image needs to be rendered at different times because the wings are too tall.
You get what I mean
so I need to put one image ABOVE the other, and have them fit together. So what you said will work?
I don't understand why you say you are going to lose a lot of detail. All you need to do is increase the number of pixels in your Render Settings tab.
Are you trying to render one scene over a background image?
Are you trying to render one scene over a background image?
No.
Okay, look, this is what I did.
Idk if this is considered nudity but it still didn't work right.
IT was still slightly uneven and I had to edit the image, like I had to edit the image with the healing brush to get rid of weird lines and the wings not ending up even.
I tried doing the merge image thing but it got all weird. It didn't matcdh up, weird areas were transparent, pieces were missing...
So I had to use the larger image, and just take the two image I rendered and put them on the image.
And as I said, it still didn't line up right...
Two images to make one bigger one. Like, the art I am doing right now involves wings, and I want to do the entire wings, but the image needs to be rendered at different times because the wings are too tall.
You get what I mean
so I need to put one image ABOVE the other, and have them fit together. So what you said will work?
Yep. I've used Photomerge to merge a panorama made up of about 5 or 6 shots of a massive old class photo that had scratches, tears, and creases in it. I can't even tell where the seams are in some areas.
It takes some getting used to, and it may take a couple tries until you get settings that work best for you, but it's completely doable.
/edit
A second option is to move your camera only on the y-axis so your renders have the same left/right edges. When you open Photoshop, create a new image that is the width of your image, but twice the height of one of your images. Place both shots on their own layer, and use shift+mouse click to drag them around on the y-axis only. From there, mask out the seams on the overlapping image until you get a perfect merge. This method is probably easier than figuring out Photomerge.
Sorry iDiru... but your description of what you are trying to achieve and how you are going about it is still confusing.
To me it reads as if you have tried to render two images as follows...
-- The 'first' image is a rendering of only the bottom half of the total image you are trying to create (the lower wings and figure).
-- The 'second' image is a rendering of the top half of the total image you are trying to create (the upper wings).
You are then patching the two smaller images together to create the larger image.
Is this correct?
If this is the case, then do not translate (move) the camera up and down to divide your image into two separate renders. Zoom in and rotate the camera around the X axis (tilt it) to capture top and bottom halves of the image. You must keep the camera's xyz coordinates, zoom, focal distance, etc. all the same for all the images you wish to patch together so that they all share the same perspective. Only change the camera rotation.
Now, the reason you divided the image up into separate renders is so that you could use different resolutions, and capture more detail in one image over another?
You lost me with this. I don't understand why you think you can't capture the resolution you need in one rendering? If your system is up to it, all you have to do is use the Render Settings to set large enough image dimensions, measured in pixels, to give you all the detail you want. The more pixels you assign the more detail your render will capture. Are you running into some size limit or some other limitation?
Or did I completely misinterpret what you are trying to do?
Signed - Curious & Confused
Yep. I've used Photomerge to merge a panorama made up of about 5 or 6 shots of a massive old class photo that had scratches, tears, and creases in it. I can't even tell where the seams are in some areas.
It takes some getting used to, and it may take a couple tries until you get settings that work best for you, but it's completely doable.
/edit
A second option is to move your camera only on the y-axis so your renders have the same left/right edges. When you open Photoshop, create a new image that is the width of your image, but twice the height of one of your images. Place both shots on their own layer, and use shift+mouse click to drag them around on the y-axis only. From there, mask out the seams on the overlapping image until you get a perfect merge. This method is probably easier than figuring out Photomerge.
Yeah, that's the thing, I only moved it on the Y-axis- that's the one that's up and down, right? I can't remember... But still somehow something went wrong SOMEWHERE. If the wing was even one one part, it was uneven on the other. I don't understand what happened...
This happened to me once in another image...
I tried rendering a woman's boots because they wouldn't fit in the pic. I used the y-axis paremeter and went down and rendered from that area. And when I tried to merge them, somehow things would always be off....
I did an area render, and I rendered parts that were already rendered to make sure that there wouldn't be anything missing. But it didn't work, still..
I included the image I worked with; I had to cut out the boots, because I just couldn't get things to line up. Like if I got her boots to line up to her legs somehow the beams in the background would be way out of alignment.
I like to render my images with their backgrounds intact because tbh I am not very good with this program. There are things that I don't understand and so I use this to make sure shadows and lighting is all equal so if I put it together in another program, it won't look weird.
She has a slight problem with her right eye, for some reason, because even though I did an area render of the exact part on her eye it didn't go to the right spot... I even did it before I moved the camera down. There was some kind of weird texture error and I decided to remove what was causing it and just render her eye.
Sorry there's so much explanation here, I'm just trying to figure out why my stuff isn't working.
I actually went back into PS and redid the wing image, to show you exactly the problem I am having.
This is what happens when I layer the images ontop of eachother...
And the exact images I rendered, not put together.
My problem is that I don't know the exact settings to set. I know I want the top of the wings visible, but I don't know how much I need to go up, number wise, to render it..
I dont' want my pic to look really small, is the problem. What I mean by losing detail is that you will see less detail, since it is small. Like if you render up close you are going to get more detail, but if you render further away you will miss some of the finer details. So I wanted the image to be large enough to be able to see everything.
This might be the critical phrase. Did you move the camera vertically (Y Translate on the Parameters tab) or rotate the camera vertically (X Rotate on the Parameters tab)? The mismatch you see appears to be because you've moved the camera, so it sees the wings at a different angle between the two renders. When you do multiple renders like this and stitch them together, it is vital that the camera must always look at the figure from the same position; you don't move it to see the rest of your figure, you turn it to see the rest of your figure. This is the same principle that lets a panoramic camera take a photo that's much wider than the camera lens can actually see.
Hold on. As I said above I have more info coming pertaining to this part of your post as well, with some info on setting up your rendering. ;-)
This might be the critical phrase. Did you move the camera vertically (Y Translate on the Parameters tab) or rotate the camera vertically (X Rotate on the Parameters tab)? The mismatch you see appears to be because you've moved the camera, so it sees the wings at a different angle between the two renders. When you do multiple renders like this and stitch them together, it is vital that the camera must always look at the figure from the same position; you don't move it to see the rest of your figure, you turn it to see the rest of your figure. This is the same principle that lets a panoramic camera take a photo that's much wider than the camera lens can actually see.
That's what I said. *rolls eyes* ;-P
This might be the critical phrase. Did you move the camera vertically (Y Translate on the Parameters tab) or rotate the camera vertically (X Rotate on the Parameters tab)? The mismatch you see appears to be because you've moved the camera, so it sees the wings at a different angle between the two renders. When you do multiple renders like this and stitch them together, it is vital that the camera must always look at the figure from the same position; you don't move it to see the rest of your figure, you turn it to see the rest of your figure. This is the same principle that lets a panoramic camera take a photo that's much wider than the camera lens can actually see.
Translate, not rotate. I simply moved the camera up. As in I moved it's location. I did not just rotate it
Straight up and down, not side to side or in and out.
I'm sorry, I don't know exactly which parameters these are, off hand. Just that I moveed it UP AND DOWN explicitly.
Yes, that's exactly what's causing your problem. You moved the camera, so it isn't seeing the wings from the same angle. You must only rotate the camera.
As for seeing the parameters, can you select the Parameters tab then select the camera in the Scene tab? That will let you see the parameters you need to adjust.
Translate, not rotate. I simply moved the camera up. As in I moved it's location. I did not just rotate it
Straight up and down, not side to side or in and out.
I'm sorry, I don't know exactly which parameters these are, off hand. Just that I moveed it UP AND DOWN explicitly.
We are saying "Do Not" move the camera. Do not change its X, Y, or Z coordinates. This means you do not want to change the X, Y or Z Translate sliders/dials.
Only rotate the camera using the X, Y or Z Rotate sliders/dials.
Make sense?
Greetings,
Wait, what?
Set your render resolution to 1081 x 1386 and render it once. Don't mess around with trying to move the camera and merge the images, just render it once at the full height resolution. If you're using DAZ Studio, that's in the Render Settings. Don't let the ratio mess you up, enter the resolution you want in as the ratio and the resolution. You should probably also turn on the render guide line display so you can see what's going to be rendered, and what's going to be cut off. (I don't recall where that setting is, and I'm mid-render right now.)
-- Morgan
Whoops.. the y-translate idea was my bad. I've taken panoramas before and not once stopped to think that I don't step sideways (x-translate), but rather turn in place (y-rotate).
@ iDiru... OK, I don't know what software you are using or how knowledgeable and experienced you are with it, but for convienence I'm going to assume you are using DAZ Studio and not Poser, and are a novice. No offense intended.
With that said, PLEASE try the following as a solution to your rendering problem.
Within DAZ Studio click on the "Viewport Options Button" in the upper right corner of the "Viewport" to display the "Viewport Options Menu" (see section 2.4.3 of the current DAZ Studio 4 Manual), and click the "Show Aspect Frame" option to turn the "Aspect Frame" on. By default the Aspect Frame is turned off. The Aspect Frame appears as a white outlined rectangle in the Viewport and shows what will be rendered in your scene and what will not be rendered. Everything inside the Aspect Frame rectangle will be rendered and everything outside it will not.
You can control the width and height dimensions of the "Aspect Frame" using the settings found within the "Dimensions" section of the "Render Settings Pane (Tab)". I recommend selecting "Letter (17:22)" for your "Preset" under the Dimensions section. Note that the "Pixels" and "Aspect" settings will automatically update to correspond with whatever Preset you choose. If the "Letter" prest doesn't give you an aspect ratio you like try one of the others or create a custom preset.
Now that you have selected an Aspect ratio for your render it is important to select the "Pixels" dimensions for the image to be sure it is rendered with the appropriate amount of detail. The more pixels the more image detail, the longer the render times, and the greater the file size. May I suggest 2550 x 3300 pixels or smaller? Why? The "Letter" aspect ratio is generally based on 8.5 x 11 inch (22 x 28 cm) paper, and full color artwork is generally printed at 300 to 600 dots/pixels per inch (118 to 236 dots/pixels per cm), and therefore... an 8.5 x 11 inch image printed at 300 pixels per inch equals 2550 x 3300 pixels. (Note: For reference, and at the risk of confusing things, images used on the Internet are usually given a resolution of 72 to 150 pixels per inch. This just means that far less pixels are used in images to be viewed via the Internet as opposed to those to be printed. This makes for smaller and quicker downloading files since there generally isn't a need for the greater image detail on the Internet as opposed to the detail needed for print.)
If you have followed the above instructions you should be able to render your entire image in one go, without the need for pasting together partial renders. Simply setup your "Aspect Frame" as described and compose your scene within its confines and render.
OK, but what if you still find a reason to want to render your image in two parts? Then do everything I just described above, including composing your scene at its final size within the Aspect Frame". Then... Change your Aspect frame so that it is a little more then half the size of your finished image. Don't move or otherwise change your camera except to rotate it down to fit the bottom half of your image within the Aspect Frame. Render. Then rotate the camera up to fit the top half of your image in the Aspect Frame and Render. If you do this you should end up with two halves of your final image that line up and overlap each other slightly when brought together in your image editing software of choice.
Questions?
Easy mistake to make. :)
~~~
@ iDiru... I just noticed your footnote about being a Poser Pro 2012 user. *rolls eyes*
The instruction I gave above for DAZ Studio would similarly apply to Poser. I'll see if I can't open up my copy of Poser and translate the instruction given for use in Poser.
I strongly suggest you just do the full render at once option as posted. The Background image you are using is not going to track properly with a camera translate. A Image on a Plane, a image loaded to the Background setting or one on a sphere or sky dome is still going to be fixed in the 3D Viewport and not be lined up properly when you only do a camera Track or translate.
From previous posts, the OP is using Poser. IIRC its 2012. As posted, the only thing the OP really needs to do is set up the render dimensions and use the render guides to see what will appear in the finished render.
Ahhh... I see I'm not the only one who missed the fact that iDiru stated in his/her signature that they are using Poser Pro 2012. So easy to miss this bit of info in the signature. ;-)
Exactly!
My question, though, is... is this enough information for iDiru to figure out what he/she needs to do, to make the intuitive leap, and to solve the problem in the simplest fashion possible? Hope so...