hey there fellows.
I'm stumbling into so many tutorials for creating 3D mashes from specific grayscale images, but I'm having a hard time finding how those images are made - some clever site says, maybe ironically, "they are made from 3D models".
Is there any tutorial on how to create 3D grayscale images from a picture?
> 3d mashes
sorry for typo, it's meshes. I know.
no you fucking retard
anyone knows how to get a usable 3D grayscale image from a standard image?
>pic related for example
The simple answer is, you can't.
Those greyscale images are heightmaps, which means that the different values represent height, where black = no elevation and white = maximum elevation.
A regular image really has no clear height data. It contains light reflected in all kinds of directions, and the different values do not represent height.
The complicated answer is that in some cases it can be approximated or faked.
There are some programs out there, like CrazyBump or AwesomeBump for example, that use some tricks to get height data from an image, but this works best for flat surfaces that does not have very complex lighting info, like brick walls, stucco surfaces, plain paper or fabric, etc.
probably the only way to draw such images is to pull up your sleeves and do a paintjob in PS/Gimp but at most won't end up anything "good".
or the good old 3d modeling.
Well, I forgot to mention that in most cases they are indeed created from 3d models. It is the best way to ensure that you get the correct height data. It is a way of getting the detail and quality of a 3d mesh without actually needing the mesh.
A height map can, for example, be used in games for parallax mapping or displacement mapping.
aside from games, why would you need a height image from a 3D model, just to use to make *another* 3D model?
actually, as I am typing, came to my mind the idea that in a mesh model, for 3D printing for example, I only need to know the "skin" geometry. this would save space and minimize prototyping errors, right?
Download Awesome Bump or install Insane Bump in Gimp, the instructions about how turn images into normal map or height map are on youtube
Aside from games, it would not have so much use (unless someone can give a good example).
The biggest benefit I can think of is saving memory on disk.
The essence of a heightmap is that it is a way of simplifying geometry data.
This means that instead of requiring a file to remember a huge load of vertex data, for example a huge landscape, detailed walls, ground or bark that consists of millions of polygons, the file would only need to save a simple mesh that can be tessellated (subdivided) and a texture.
I would not recommend using a heightmap for 3d printing. Unless someone can tell me it is a common practice, it would be ugly and possibly risky, due to it's nature.
today a customer walked in and asked for a ring with his "heraldic" crest (what a moron - but ok, it's work) on it, and gave me a hilriously big .tiff file (19MB) with an engraving drawing he used to make his stamp.
on a 3D book I used at school (Rhino3D, McNeel) I found this nice ZSurf command... but the book didn't tell from where the grayscale images came from. I thought there was some command in PS or similar to get them from images...
by any chance... how the heck one can get a heightmap image from a, let's say, .stl file?
I Googled it quickly, but Zsurf would have the same problem. You can use any image, but as long as that image does not have proper height data, it will always look strange.
I've attached a picture of a height map I've made a few years ago for a game I was working on at the time. I don't think it actually ended up in the game, so I think it is safe to show here.
I made this heightmap in Photoshop by using separate layers that had the 'inner glow' and 'colour overlay' layer effects applied. this allowed me a lot of control and guaranteed relative smoothness of the gradients.
Heightmaps are greyscale images, where black = no elevation and white = maximum elevation. so you would need to reproduce that effect in the 3d software. There are several ways of doing that.
- use a shader that has a gradient along the axis that points towards the camera, so that the darkest value is furthest away from the camera and white the closest. (easiest)
- use a linear fog or mist, whereby the visible limits match the range of the 3d object you are using.
- use a non-shadowing directional light that casts light (with linear falloff) along the axis going away from the camera.
very interesting. I'll try this!
uff. I forgot that Vray didn't come bundled with rhino. and Rhino may be good for working but it's pretty low level in terms of rendering. sigh.
sorry anon, I realize posted the wrong image.
This is the actual height map image I used.
oh, thank you very much!