Re: [algogeeks] image processing

yes it is matching points between 2 different pictures of the same object taken in different environments..not different views of the same object On Sun, Jun 26, 2011 at 5:57 PM, sukhmeet singh wrote: > YES.. we need to have 2 shots.. !! > u said u had some matching points in the 2 images >

Re: [algogeeks] image processing

YES.. we need to have 2 shots.. !! u said u had some matching points in the 2 images On Sun, Jun 26, 2011 at 6:23 PM, Arun Vishwanathan wrote: > hmm for starting this , we need 2 shots of a scene right? I have only a > single image..no stereo vision here > > > On Sun, Jun 26, 2011 at 2:16 PM

Re: [algogeeks] image processing

hmm for starting this , we need 2 shots of a scene right? I have only a single image..no stereo vision here On Sun, Jun 26, 2011 at 2:16 PM, sukhmeet singh wrote: > I don't know how far u have been able to conquer the problem .. We tried > this and used epipolar geometry and 7point algo and found

Re: [algogeeks] image processing

I don't know how far u have been able to conquer the problem .. We tried this and used epipolar geometry and 7point algo and found the Homography matrix.. try referring to Hartley and Zeisserman book.!! On Thu, Jun 23, 2011 at 3:52 PM, DK wrote: > Perspective transformations are non linear becau

Re: [algogeeks] image processing

Perspective transformations are non linear because in this case Y = AX the matrix A is a function of Y and X. It would be linear only if A was independent of X and Y. (See the derivation on that link). -- DK http://twitter.com/divyekapoor http://www.divye.in -- You received this message becaus

Re: [algogeeks] image processing

Thanks DK! I have a kinda small doubt.Generally when u say a transformation is of the form y=Ax and in this case A is the tranformation matrix it implies a linear transformation right ? we express even as such for perspective right? but perspective transformations are not linear right? I am sorry

Re: [algogeeks] image processing

Have you tried pixel interpolation? (Because affines are pretty easy and you should try really hard to make them work). If it really is a perspective transformation then check: http://alumni.media.mit.edu/~cwren/interpolator/ (I have no experience with this so you're on your own after this). --

Re: [algogeeks] image processing

thanks DK... what if its a perspective transformation and not affine exactly? cos i tried using this matrix [xnew ynew 1]=[a b c;d e f;0 0 1][x y 1] now after getting the matrix parameters using the feature points matched in both the images , when i put the border pixels as input to this matrix th

Re: [algogeeks] image processing

If the two preconditions for an affine relationship are met: The collinearity relation between points; i.e., the points which lie on a line continue to be collinear after the transformation Ratios of distances along a line; i.e., for distinct colline

Re: [algogeeks] image processing

logarithmis algorithm can be an answer to the question. On Mon, Jun 20, 2011 at 8:58 PM, Arun Vishwanathan wrote: > hi guys, > > I need some helpgiven 2 images and the matching feature points in the 2 > images( maybe some false matches but rare), how can I get the transformation > matrix from

[algogeeks] image processing

hi guys, I need some helpgiven 2 images and the matching feature points in the 2 images( maybe some false matches but rare), how can I get the transformation matrix from this so that any other point in source image can be almost correctly translated to a point in destination image?? does anybo