compute epipolar lines on the image.

Epipolar line is calculated by "l = F * p" if which_image is 0 or "l = F^T * p" if which_image is 1. The line is represented as "a*x + b*y + c = 0", where "l = [a, b, c]^T".

引数

[in]	points	is a array of points. The size is 2xN or 3xN.
[in]	which_image	is an index of the image including the points (0 or 1).
[in]	fundamental	matrix is a input fundamental matrix.
[out]	correspondent_lines	are epipolar lines which corresponds to the points. The size is 3xN.

戻り値

return true if success

compute epipole from a fundamental matrix

引数

[in]	fundamental	matrix is a input fundamental matrix.
[out]	epipole	is a (3, 1) matrix

戻り値

return true if success

参照先 mist::svd().

Compute fundamental matrices from trifocal tensor.

引数

[in]	tensor	is the input trifocal tensor
[out]	fundamental21	is the fundamental matrix with the second image (3x3)
[out]	fundamental31	is the fundamental matrix with the third image (3x3)

戻り値

Return true if success

参照先 mist::matrix< T, Allocator >::resize(), と trifocal_tensor_to_epipoles().

compute a fundamental matrix from 8 or more correspondence of the points.

The fundamental matrix is computed using 8 point algorithm.

引数

[in]	points1	is an array of points on image1. The size is 2xN or 3xN.
[in]	points2	is an array of points on image2. The size is 2xN or 3xN.
[out]	fundamental_matrix	is a fundamental matrix. The size is 3x3.

戻り値

return true if success

参照先 _DESCENDING_ORDER_EIGEN_VALUE_, と mist::eigen().

Compute trifocal tensor from trilinear relations.

Solve At=0 using SVD, where t is the vector of the trifocal tensor components and A are the coefficients of the trilinear relations.

引数

[in]	relations	is the trilinear relations (Nx27)
[out]	tensor	is the output trifocal tensor

戻り値

Return true if success

参照先 mist::matrix< T, Allocator >::cols(), mist::matrix< T, Allocator >::rows(), と mist::svd().

factorize essential matrix

factorize E to SR, where S = U Z U^t R1 = U W V^t or R2 = U W^t V^t E = U diag(1, 1, 0) V^t W = { (0,1,0)^t, (-1,0,0)^t, (0,0,1)^t } Z = { (0,-1,0)^t, (1,0,0)^t, (0,0,0)^t }

引数

[in]	essential_matrix	is a input
[out]	t	is the traslation vector
[out]	r1	is one of the solution of the rotation R
[out]	r2	is the other solution of the rotaion R

戻り値

return true if success

参照先 mist::matrix< T, Allocator >::_33(), mist::svd(), と mist::matrix< T, Allocator >::t().

Estimate a coordinate of a point in world coordinate. The triangulation is performed by DLT method.

引数

[in]	xc1	is the point on the first camera
[in]	xc2	is the point on the second camera
[in]	camera_matrix1	is that for the first camera
[in]	camera_matrix2	is that for the second camera
[out]	xw	is the point on the world coordinate

戻り値

return true if success

参照先 mist::svd().

Compute camera matrices from trifocal tensor.

The first camera matrix is fixed as P = [I | 0]. The camera matrices are obtained up to projection ambiguity.

引数

[in]	tensor	is the input trifocal tensor
[out]	mat2	is the camera matrix for the second image (3x4)
[out]	mat1	is the camera matrix for the third image (3x4)

戻り値

Return true if success

参照先 mist::matrix< T, Allocator >::resize(), mist::matrix< T, Allocator >::t(), と trifocal_tensor_to_epipoles().

Compute epipoles from trifocal tensor.

引数

[in]	tensor	is the input trifocal tensor
[out]	epi21	is the epipole in the second image (3x1)
[out]	epi31	is the epipole in the third image (3x1)

戻り値

Return true if success

参照先 mist::matrix< T, Allocator >::resize(), と mist::svd().

参照元 compute_fundamental_matrices(), と trifocal_tensor_to_camera_matrices().

Generate trilinear relations from 3-view line-line-line correspondence.

Return A, where At=0 and t is the vector of the trifocal tensor components.

引数

[in]	l1	is a line on the first camera (3x1)
[in]	l2	is a line on the second camera (3x1)
[in]	l3	is a line on the third camera (3x1)

戻り値

trilinear relations (3x27)

Generate trilinear relations from 3-view point-point-pont correspondences.

Return A, where At=0 and t is the vector of the trifocal tensor components.

引数

[in]	p1	is a point on the first camera (3x1)
[in]	p2	is a point on the second camera (3x1)
[in]	p3	is a point on the third camera (3x1)

戻り値

Return trilinear relations (9x27)