Revision history [back]

I'm not sure if this is trivial in OpenCV and c++, but after looking vastly around on websites, I still didn't find an elegant way to solve this.

So I have 3x3 camera matrix K of type Matx33f and #N-homogenous image points stored in a vector<point3f> named X. What I'd like to achieve is the calculation of the projective rays via:

Xn = K.inv() * X_vec,

whereby X_vec = [x1 x2 ... xN; y1 y2 ... yN; 1 1 ... 1]; (3xN)

I know how to achieve this with a loop, but I wonder if there isn't a simpler and elegant way. Can't imagine it to be that costly.

Many thanks in advance.

I'm not sure if this is trivial in OpenCV and c++, but after looking vastly around on websites, I still didn't find an elegant way to solve this.

So I have 3x3 camera matrix K of type Matx33f and #N-homogenous image points stored in a vector<point3f> named X. What I'd like to achieve is the calculation of the projective rays via:

Xn = K.inv() * X_vec,

whereby X_vec = [x1 x2 ... xN; y1 y2 ... yN; 1 1 ... 1]; (3xN)

Edit: The problem is that X is of type Point2f and I need a vector form in the way of X_vec. I know how to achieve this to the muliplication with a loop, but I wonder if there isn't a simpler and elegant way. way to transform Point2f X into X_vec. Can't imagine it to be that costly.

Many thanks in advance.

I'm not sure if this is trivial in OpenCV and c++, but after looking vastly around on websites, I still didn't find an elegant way to solve this.

So I have 3x3 camera matrix K of type Matx33f and #N-homogenous image points stored in a vector<point3f> named X. What I'd like to achieve is the calculation of the projective rays via:

Xn = K.inv() * X_vec,

whereby X_vec = [x1 x2 ... xN; y1 y2 ... yN; 1 1 ... 1]; (3xN)

Edit: The problem is that X is of type Point2f and but I need it in a vector form in the way of X_vec. I know how to to the muliplication with a loop, but I wonder if there isn't a simpler and elegant way to transform Point2f X into X_vec. Can't imagine it to be that costly.

Many thanks in advance.

I'm not sure if this is trivial in OpenCV and c++, but after looking vastly around on websites, I still didn't find an elegant way to solve this.

So I have 3x3 camera matrix K of type Matx33f and #N-homogenous image points stored in a vector<point3f> named X. What I'd like to achieve is the calculation of the projective rays via:

Xn = K.inv() * X_vec,

whereby X_vec = [x1 x2 ... xN; y1 y2 ... yN; 1 1 ... 1]; (3xN)

Edit: The problem is that X is of type Point2f Point3f and but I need it in a vector form in the way of X_vec. I know how to to the muliplication with a loop, but I wonder if there isn't a simpler and elegant way to transform Point2f X into X_vec. Can't imagine it to be that costly.

Many thanks in advance.

I'm not sure if this is trivial in OpenCV and c++, but after looking vastly around on websites, I still didn't find an elegant way to solve this.

So I have 3x3 camera matrix K of type Matx33f and #N-homogenous image points stored in a vector<point3f> named X. What I'd like to achieve is the calculation of the projective rays via:

Xn = K.inv() * X_vec,

whereby X_vec = [x1 x2 ... xN; y1 y2 ... yN; 1 1 ... 1]; (3xN)

Edit: The problem is that X is of type Point3f and but I need it in a vector form in the way of X_vec. I know how to to the muliplication with a loop, but I wonder if there isn't a simpler and elegant way to transform Point2f X into X_vec. Can't imagine it to be that costly.

Many thanks in advance.

I'm not sure if this is trivial in OpenCV and c++, but after looking vastly around on websites, I still didn't find an elegant way to solve this.

So I have 3x3 camera matrix K of type Matx33f and #N-homogenous image points stored in a vector<point3f> named X. What I'd like to achieve is the calculation of the projective rays via:

Xn = K.inv() * X_vec,

whereby X_vec = [x1 x2 ... xN; y1 y2 ... yN; 1 1 ... 1]; (3xN)

Edit: The problem is that X is of type <vector point3f="" &gt;="" but="" i="" need="" it="" in="" a="" vector="" form="" in="" the="" way="" of="" x_vec.="" i="" know="" how="" to="" to="" the="" muliplication="" with="" a="" loop,="" but="" i="" wonder="" if="" there="" isn't="" a="" simpler="" and="" more="" elegant="" way="" to="" transform="" <vector="" point3f="" &gt;="" x="" into="" x_vec.="" can't="" imagine="" it="" to="" be="" that="" costly.<="" p="">

Many thanks in advance.

I'm not sure if this is trivial in OpenCV and c++, but after looking vastly around on websites, I still didn't find an elegant way to solve this.

So I have 3x3 camera matrix K of type Matx33f and #N-homogenous image points stored in a vector<point3f> named X. What I'd like to achieve is the calculation of the projective rays via:

Xn = K.inv() * X_vec,

whereby X_vec = [x1 x2 ... xN; y1 y2 ... yN; 1 1 ... 1]; (3xN)

Edit: The problem is that X is of type <vector point3f=""> vector<point3f> but I need it in a vector form in the way of X_vec. I know how to to the muliplication with a loop, but I wonder if there isn't a simpler and more elegant way to transform <vector point3f=""> vector<point3f> X into X_vec. Can't imagine it to be that costly.

Many thanks in advance.