Sign in to follow this  
MohammadAhmed

computer vision

Recommended Posts

jwezorek    2663

The phrase "computer vision" much like "artificial intelligence" is not a well defined term and tends to me different things to different people.

 

Could you be more specific about what you want to learn about?

Share this post


Link to post
Share on other sites
MohammadAhmed    279

@jwezorek,

I need books for computer vision related to 3D associated with math books for it smile.png .However , I don't know if there is field in computer vision related to 3D games????

by the way, I don't want computer vision books for 
robot or digital camera no but for 3D (3D computer graphics ) of course smile.png.

I don't know much about this field !! so I don't know if my question is right ! 

 

I just know about computer graphics and digital image processing.
I took as prerequisites math books ( Calculas 1  & 2 , linear algebra , discrete math ).

 

Edited by mrmohadnan

Share this post


Link to post
Share on other sites
alh420    5995
"Computer Vision" usually means making the computer understand and interpret signals, from for example cameras or images (typically to build an awareness or internal model of its content) so I don't really understand what you mean with computer vision for 3D and games...

Do you mean visualisation? That is, how to generate an image from your 2D or 3D data? Rendering, Raytracing, things like that? Edited by Olof Hedman

Share this post


Link to post
Share on other sites
jwezorek    2663
mrmohadnan I agree with what Olof says above. I'm not sure what you mean by computer vision applied to 3D games. Could you tell us more about what you want to do? and we'll try to tell you what to research. Edited by jwezorek

Share this post


Link to post
Share on other sites
jwezorek    2663

I think as an introduction '3D Math Primer for Graphics and Game Development' might suit.

Well, yeah, that's about 3D math and graphics but the the OP was asking about computer vision.

I'm not sure what the OP means. The only thing I can think of is that "computer graphics" is basically about doing 3D model -> 2D raster image. Maybe mrmohadnan wants to go 2D raster image -> 3D model -- that is "computer vision" and is sort of related to 3D games, but without more info not sure. Edited by jwezorek

Share this post


Link to post
Share on other sites
MohammadAhmed    279

@Olof Hedman , jwezorek , eppo ,

thanks , 

so you say visualisation (rendering , ray...etc) are different from Computer Vision field  , and they don't relate?

Q1 : Okay 
Olof as you say , I have to take signal processing course and random theory as prerequisites for Computer vision right?

jwezorek , 

I wanna to take data from Camera and capture objects with accosiated with it's cloth ( Material ) and import them into 3D max and export it as Model ready for game engine without needs to make materials in 3d application smile.png (the problem I don't know anything about signal processing !!!!! because Olof says  ,"understand and interpret signals " :(

 

 


Q2 : So thats why I am asking , is there a field in computer vision close to Visualization (Rendering , tracing ...etc) or not ?

Edited by mrmohadnan

Share this post


Link to post
Share on other sites
jwezorek    2663

I wanna to take data from Camera and capture objects with accosiated with it's cloth ( Material ) and import them into 3D max and export it as Model ready for game engine without needs to make materials in 3d application

I don't know ... but I'm afraid extracting 3D models and material information from a 2D image in a completely unsupervised manner is beyond the state of the art of computer vision today. Unless I'm misunderstanding...

Share this post


Link to post
Share on other sites
alh420    5995

Actually, you can do it, if you have several images of the object from different angles.

But it's a very advanced subject, and you will not get models you can use directly in a game.

 

One low cost solution for a simple "3D scanner" is to use one (or two for better accuracy) kinects. 

The API is open, and you get 2 images, one normal camera image, and one depth image, with the distance from the kinect to each point in the camera image.

I'm pretty sure you can find some software to do it out there, at least there are for motion capture.

But any model extracted from it would also be very noisy and need extensive cleanup to be usable in a game.

 

But it seems it is computer vision you want, after all.

I don't have any tips for any books unfortunately...

Share this post


Link to post
Share on other sites
MohammadAhmed    279
jwezorek

you are right :) that's what I want from "2D -> 3D" 

I used google and found that Computer vision in general the mathmematics requirements are :
geometry ( any recommended books ?)
Optimiazation ( what you prefer for Optimization : " linear programming " , "complex analysis ")
statistics ( I think introduction to probability course is good !)
and about physics prerequisities books for computer vision :
I think that digital signal processing is enough !

Share this post


Link to post
Share on other sites

Create an account or sign in to comment

You need to be a member in order to leave a comment

Create an account

Sign up for a new account in our community. It's easy!

Register a new account

Sign in

Already have an account? Sign in here.

Sign In Now

Sign in to follow this