Right hand camera to left hand, OpenCV to Unity - opencv

I have a problem that has been puzling me for the last few days. I have a camera pose obtained with Opencv that is right handed (X-right, Y-up, Z-back) and I would like to visualize in Unity (X-right, Y-up, Z-forward) but I cannot really manage to get it right.
I tried to do it using either quaternions or matrices and it should be just mirroring the Z axis and set the rotation of the transform of the camera in Unity to the computed transformation however I cannot get the right conversion.
With quaternions I tried to mirror by negating the Z and W term and I achieved a coordinate system (X-right, Y-down, Z-forward), it makes sense but it is not what I want to achieve. With matrices I think I should multiply my right hand camera by an identity matrix with the element [2,2] set to -1, however I don't get what I want.
I am definitely missing something, probably something really stupid I forgot :)
Has anybody a suggestion?

A Quaternion can be thought of as a rotation around an axis a = (ax, ay, az) by an angle theta
qx = ax * sin(theta/2)
qy = ay * sin(theta/2)
qz = az * sin(theta/2)
qw = cos(theta/2)
In a right-handed coordinate system a rotation of theta will be counter-clockwise, while in a left-handed coordinate system a rotation of theta will be clockwise (depending on your point-of-view, of course).
So to get your quaternion from a right-handed system to Unity's Left-Handed system you have to account for two factors:
The Z-Axis is negated
The direction of rotation is flipped from CCW to CW
We first factor is accounted for by negating the qz component of the quaternion. The second factor is accounted for by flipping the axis of rotation (rotating by 90 degrees around 1,0,0 is the inverse of rotating 90 degrees around -1,0,0).
If your original right-handed quaternion is q and your left-handed quaternion is q'That means you end up with:
q'=(-qx, -qy, qz, qw)
Additional Note
Quaternions don't inherently have a handedness. A quaterion q applies equally well in RH or LH coordinate systems. However, when you apply the quaternion to a spatial vector, the resulting transformation takes on the handedness of the vector's space.

Related

How can we find theta from rotation matrix?

According to OpenCV's documentation, solvePnp will return the rotation vector of the object pose from 3D-2D point correspondes. To obtain the rotation matrix, we can use Rodrigues method to convert the rotation vector to rotation matrix. According to OpenCV documentation, we can find theta using the following:
theta = norm(r)
But I thought norm(r) will find the magnitude of the vector r? If that's the case how can we find the angle from the magnitude of the vector r? Correct me if I am wrong. Thank you.
Given a rotation vector r, its length (in Python, numpy.linalg.norm(r)) is the angle of rotation around the axis whose direction is the vector's one. The sense of the rotation obeys the "right hand rule": if your right hand makes a thumbs-up sign, with the thumb pointing as the vector, the other fingers wrap the vectors as the rotation (equivalently, it's the sense of rotation that makes an ordinary screw advance when its tip points as the vector).
The same rotation can be expressed as a 3x3 matrix, or as a triple of (Euler) angles or rotation about up to 3 orthogonal axes. There are ordinarily many different triples of Euler angles that represent the same rotation. Consult a textbook, or Wikipedia, for details.

Opencv get R and t from Essential Matrix

I'm new to opencv and computer vision. I want to find the R and t matrix between two camera pose. So I generally follows the wikipedia:
https://en.wikipedia.org/wiki/Essential_matrix#Determining_R_and_t_from_E
I find a group of the related pixel location of the same point in the two images. I get the essential matrix. Then I run the SVD. And print the 2 possible R and 2 possible t.
[What runs as expected]
If I only change the rotation alone (one of roll, pitch, yaw) or translation alone (one of x, y, z), it works perfect. For example, if I increase pitch to 15 degree, then I would get R and find the delta pitch is +14.9 degree. If I only increase x for 10 cm, then the t matrix is like [0.96, -0.2, -0.2].
[What goes wrong]
However, if I change both rotation and translation, the R and t is non-sense. For example, if I increase x for 10 cm and increase pitch to 15 degree, then the delta degree is like [-23, 8,0.5], and the t matrix is like [0.7, 0.5, 0.5].
[Question]
I'm wondering why I could not get a good result if I change the rotation and translation at the same time. And it is also confusing why the unrelated rotation or translation (roll, yaw, y, z) also changes so much.
Would anyone be willing to figure me out? Thanks.
[Solved and the reason]
OpenCV use a right-hand coordinate system. This is to say that the z-axis is projected from xy plane to the viewer direction. And our system is using a left-hand coordinate system. So as long as the changes are related to the z-axis, the result is non-sense.
This is solved due to the difference of the coordinates using.
OpenCV use a right-hand coordinate system. This is to say that the z-axis is projected from xy plane to the viewer direction. And our system is using a left-hand coordinate system. So as long as the changes are related to the z-axis, the result is non-sense.

How to calculate camera orientation using one point in large distance (using opencv)?

Let's say I have a pinhole camera with known intristic values like camera matrix and distortion coefficients. Let's say there is a point in large enough distance from the camera, so we can say it is placed in infinity.
Given image coordinates of this point in pixels, I would like to calculate camera rotation relative to the axis that connects camera and this point (so rotation is 0,0 if camera is directed at this point and it is in the optical center of the image).
How can this be done using opencv?
Many thanks!
You need to specify an additional constraint - rotating the camera from its current pose to one that aligns the optical axis with an arbitrary ray leaves the camera free to rotate about the ray itself (i.e. it leaves the "roll" angle unspecified).
Let's assume that you want the roll to be zero, i.e. that you want the motion to be a pure pan-tilt. This has a unique solution as long as the ray you want to align to is not parallel to the vertical image axis (in which case pan and roll are the same motion).
Then the solution is computed as follows. Let's use the OpenCV camera frame: Z=[0,0,1]' (, where " ' " means transpose) be the camera focal axis, oriented going out of the lens, Y=[0,1,0]' the vertical axis going down, and X = Z x Y (where 'x' is the cross product) the horizontal camera axis going toward the right of the image. So "pan" is a rotation about Y, "tilt" is a rotation about X.
Let U = [u1, u2, u3]', with || u || = 1 be the ray you want to rotate to. You want to apply a pan that brings Z onto the plane Puy defined by the vectors u and Y, then apply a tilt that brings Z onto u.
The angle of the first rotation is (angle between Z and Puy) = [90 deg - (angle between Z and Y x U)]. this is because Y x U is orthogonal to Puy. Look up the expressions for computing the angle between vectors on Wikipedia or elsewhere online. Once you have the angle (or its cosine and sine), the rotation about Y can be expressed as a standard rotation matrix Ry.
The angle of the second rotation, about X after once Z is onto Puy, is the angle between vector Z and U after Ry is applied to Z, or equivalently, between Z and inv(Ry) * U. Compute the angle between the vector, and use to build a standard rotation matrix about X, Rx
The final transformation is then Rx * Ry.

finding the real world coordinates of an image point

I am searching lots of resources on internet for many days but i couldnt solve the problem.
I have a project in which i am supposed to detect the position of a circular object on a plane. Since on a plane, all i need is x and y position (not z) For this purpose i have chosen to go with image processing. The camera(single view, not stereo) position and orientation is fixed with respect to a reference coordinate system on the plane and are known
I have detected the image pixel coordinates of the centers of circles by using opencv. All i need is now to convert the coord. to real world.
http://www.packtpub.com/article/opencv-estimating-projective-relations-images
in this site and other sites as well, an homographic transformation is named as:
p = C[R|T]P; where P is real world coordinates and p is the pixel coord(in homographic coord). C is the camera matrix representing the intrinsic parameters, R is rotation matrix and T is the translational matrix. I have followed a tutorial on calibrating the camera on opencv(applied the cameraCalibration source file), i have 9 fine chessbordimages, and as an output i have the intrinsic camera matrix, and translational and rotational params of each of the image.
I have the 3x3 intrinsic camera matrix(focal lengths , and center pixels), and an 3x4 extrinsic matrix [R|T], in which R is the left 3x3 and T is the rigth 3x1. According to p = C[R|T]P formula, i assume that by multiplying these parameter matrices to the P(world) we get p(pixel). But what i need is to project the p(pixel) coord to P(world coordinates) on the ground plane.
I am studying electrical and electronics engineering. I did not take image processing or advanced linear algebra classes. As I remember from linear algebra course we can manipulate a transformation as P=[R|T]-1*C-1*p. However this is in euclidian coord system. I dont know such a thing is possible in hompographic. moreover 3x4 [R|T] Vector is not invertible. Moreover i dont know it is the correct way to go.
Intrinsic and extrinsic parameters are know, All i need is the real world project coordinate on the ground plane. Since point is on a plane, coordinates will be 2 dimensions(depth is not important, as an argument opposed single view geometry).Camera is fixed(position,orientation).How should i find real world coordinate of the point on an image captured by a camera(single view)?
EDIT
I have been reading "learning opencv" from Gary Bradski & Adrian Kaehler. On page 386 under Calibration->Homography section it is written: q = sMWQ where M is camera intrinsic matrix, W is 3x4 [R|T], S is an "up to" scale factor i assume related with homography concept, i dont know clearly.q is pixel cooord and Q is real coord. It is said in order to get real world coordinate(on the chessboard plane) of the coord of an object detected on image plane; Z=0 then also third column in W=0(axis rotation i assume), trimming these unnecessary parts; W is an 3x3 matrix. H=MW is an 3x3 homography matrix.Now we can invert homography matrix and left multiply with q to get Q=[X Y 1], where Z coord was trimmed.
I applied the mentioned algorithm. and I got some results that can not be in between the image corners(the image plane was parallel to the camera plane just in front of ~30 cm the camera, and i got results like 3000)(chessboard square sizes were entered in milimeters, so i assume outputted real world coordinates are again in milimeters). Anyway i am still trying stuff. By the way the results are previosuly very very large, but i divide all values in Q by third component of the Q to get (X,Y,1)
FINAL EDIT
I could not accomplish camera calibration methods. Anyway, I should have started with perspective projection and transform. This way i made very well estimations with a perspective transform between image plane and physical plane(having generated the transform by 4 pairs of corresponding coplanar points on the both planes). Then simply applied the transform on the image pixel points.
You said "i have the intrinsic camera matrix, and translational and rotational params of each of the image.” but these are translation and rotation from your camera to your chessboard. These have nothing to do with your circle. However if you really have translation and rotation matrices then getting 3D point is really easy.
Apply the inverse intrinsic matrix to your screen points in homogeneous notation: C-1*[u, v, 1], where u=col-w/2 and v=h/2-row, where col, row are image column and row and w, h are image width and height. As a result you will obtain 3d point with so-called camera normalized coordinates p = [x, y, z]T. All you need to do now is to subtract the translation and apply a transposed rotation: P=RT(p-T). The order of operations is inverse to the original that was rotate and then translate; note that transposed rotation does the inverse operation to original rotation but is much faster to calculate than R-1.

opencv giving camera pose - rvec and tvec - with mismatched coordinate system

I need to find the pose (rotation matrix + translation vector) for a camera, and for that I am using cv2.solvePnP(), but the results I get from photos don't match.
In order to debug, I created (with numpy) a "debugging 3d scene" composed of some object points (four corners of a square), some camera points (focal point, principal point and four corners of the virtual projection plane) and parameters (focal distance, initial orientation).
Then, I construct a general rotation matrix by multiplying three axis rotation matrices, apply this general rotation to the camera (numpy.dot()), project the object points in the virtual projection plane (line-plane intersection algorithm), and calculate the in-plane 2D coordinates (point-line distance) to the projection plane axes.
After doing this (objectpoints to imagepoints via rotationmatrix), I feed the imagepoints and the objectpoints to cv2.Rodrigues(cv2.solvePnP(...)) and get a matrix "not quite identical" to the one I used, only because of transposition and some elements with opposite signal (negative vs. positive), respecting this relation:
solvepnp_rotmatrix = my_original_matrix.transpose * [ 1 1 1]
[ 1 1 -1]
[-1 -1 1]
Although the rotation matrix mismatch is "solveable" with this hack, the TRANSLATION VECTOR gives coordinates that don't make sense to me.
I suspect there are mismatches between my 3D model (handedness, axes orientation, order of rotations) and the model used by opencv:
I use OpenGL-like coordinate system (X increases to the right, Y increases upwards, and Z increases toward the observer;
I applied the rotations in the order that made more sense to me (all right-handed, first around global Z, then around global X, then around global Y);
The image plane is between object and camera focal point (virtual projection plane, instead of real/CCD);
The origin of my image plane (virtual CCD) is lower-left corner (Xpix increases to the right, Ypx increases upwards.
My questions are:
Given that the terms of the rotation matrix are identical, only transposed and with different signal in some terms, is it possible that I am confusing some of openCV conventions (handedness, order of rotations, axis direction)? And how can I discover which one(s)?
Also, is there a way to relate my handmade translation vector to the tvec returned by solvePnP? (of course, ideally, the best would be to make the coordinate systems to match, in the first place).
Any help will be most welcome!

Resources