I would like to calculate the Horizontal and Vertical field of view from the camera intrinsic matrix for the cameras used in the KITTI dataset. The reason I need the Field of view is to convert a depth map into 3D point clouds.
Though this question has been asked quite a long time ago, I felt it needed an answer as I ran into the same issue and was unable to find any info on it.
I have however solved it using the information available in this document and some more general camera calibration documents
Firstly, we need to convert the supplied disparity into distance. This can be done through fist converting the disp map into floats through the method in the dev_kit where they state:
disp(u,v) = ((float)I(u,v))/256.0;
This disparity can then be converted into a distance through the default stereo vision equation:
Depth = Baseline * focal length/ Disparity
Now come some tricky parts. I searched high and low for the focal length and was unable to find it in documentation.
I realised just now when writing that the baseline is documented in the aforementioned source however from section IV.B we can see that it can be found in P(i)rect indirectly.
The P_rects can be found in the calibration files and will be used for both calculating the baseline and the translation from uv in the image to xyz in the real world.
The steps are as follows:
For pixel in depthmap:
xyz_normalised = P_rect \ [u,v,1]
where u and v are the x and y coordinates of the pixel respectively
which will give you a xyz_normalised of shape [x,y,z,0] with z = 1
You can then multiply it with the depth that is given at that pixel to result in a xyz coordinate.
For completeness, as P_rect is the depth map here, you need to use P_3 from the cam_cam calibration txt files to get the baseline (as it contains the baseline between the colour cameras) and the P_2 belongs to the left camera which is used as a reference for occ_0 files.
Related
I've successfully calibrated my camera and I can get the dimensions of a XLD in world coordinates with ContourToWorldPlaneXld and then HeightWidthRatioXld. This returns me the measures of a contour extracted from a shape.
Now I need to convert a value inserted by the user in mm (example in mm: 0.1) and get how many pixels the measure is, for example, to draw a line.
I need the pixel value as per request. I tried looking around in the Halcon documentation but I didn't find what I was looking for.
Also I read this answer but it' not exactly what I'm looking for.
I'm using Halcon Progress 21.11.
Edit: A possible solution could be obtaining the dimensions before converting them to world plane and then do something like pixel/world but I would prefer a better method if it exists.
I'm using a Numpy implementation of camera calibration by direct linear transformation (DLT) in python.
I'm trying to use it for 3 dimensional camera calibration.
My problem is, the mean error of the DLT (mean residual of the DLT transformation in units of camera coordinates) is very high in the example, in the thousands of pixels especially compared to the examples provided by the original author (see here).
These are the 3D points I use:
objpoints = [[86.438, -174.922,51.316],[-27.519,-215.460,39.154],
[73.601, 107.800,120.455],[87.602,133.413,34.023],
[101.276,-55.204,108.884],[88.509,-68.038,116.634],
[27.518,-215.460,39.154],[-31.355,-207.334,85.184],
[87.601,-131.059,33.881],[-60.234,-23.833,148.269],[62.162,-23.042,148.715]]
These are the pixels I use:
imgpoints = [[576.0,861.0],[660.0,996.0],[253.0,1383.0],[575.0,1481.0],
[276.0,1217.0],[241.0,1139.0],[665.0,461.0],[231.0, 411.0],
[660.0,226.0],[141.0,684.0],[111.0,1123.0]]
I extracted these points manually, for 3D from a point cloud model (.ply format) and for matching 2D image by pixels.
Something must be wrong with my coordinates at a very basic level, but I'm not sure what it is and how to find it.
I have the task to simulate a camera with a full well capacity of 10.000 Photons per sensor element
in numpy. My first Idea was to do it like that:
camera = np.random.normal(0.0,1/10000,np.shape(img))
Imgwithnoise= img+camera
but it hardly shows an effect.
Has someone an idea how to do it?
From what I interpret from your question, if each physical pixel of the sensor has a 10,000 photon limit, this points to the brightest a digital pixel can be on your image. Similarly, 0 incident photons make the darkest pixels of the image.
You have to create a map from the physical sensor to the digital image. For the sake of simplicity, let's say we work with a grayscale image.
Your first task is to fix the colour bit-depth of the image. That is to say, is your image an 8-bit colour image? (Which usually is the case) If so, the brightest pixel has a brightness value = 255 (= 28 - 1, for 8 bits.) The darkest pixel is always chosen to have a value 0.
So you'd have to map from the range 0 --> 10,000 (sensor) to 0 --> 255 (image). The most natural idea would be to do a linear map (i.e. every pixel of the image is obtained by the same multiplicative factor from every pixel of the sensor), but to correctly interpret (according to the human eye) the brightness produced by n incident photons, often different transfer functions are used.
A transfer function in a simplified version is just a mathematical function doing this map - logarithmic TFs are quite common.
Also, since it seems like you're generating noise, it is unwise and conceptually wrong to add camera itself to the image img. What you should do, is fix a noise threshold first - this can correspond to the maximum number of photons that can affect a pixel reading as the maximum noise value. Then you generate random numbers (according to some distribution, if so required) in the range 0 --> noise_threshold. Finally, you use the map created earlier to add this noise to the image array.
Hope this helps and is in tune with what you wish to do. Cheers!
Recently I'm struggling with a pose estimation problem with a single camera. I have some 3D points and the corresponding 2D points on the image. Then I use solvePnP to get the rotation and translation vectors. The problem is, how can I determine whether the vectors are right results?
Now I use an indirect way to do this:
I use the rotation matrix, the translation vector and the world 3D coordinates of a certain point to obtain the coordinates of that point in Camera system. Then all I have to do is to determine whether the coordinates are reasonable. I think I know the directions of x, y and z axes of Camera system.
Is Camera center the origin of the Camera system?
Now consider the x component of that point. Is x equavalent to the distance of the camera and the point in the world space in Camera's x-axis direction (the sign can then be determined by the point is placed on which side of the camera)?
The figure below is in world space, while the axes depicted are in Camera system.
========How Camera and the point be placed in the world space=============
|
|
Camera--------------------------> Z axis
| |} Xw?
| P(Xw, Yw, Zw)
|
v x-axis
My rvec and tvec results seems right and wrong. For a specified point, the z value seems reasonable, I mean, if this point is about one meter away from the camera in the z direction, then the z value is about 1. But for x and y, according to the location of the point I think x and y should be positive but they are negative. What's more, the pattern detected in the original image is like this:
But using the points coordinates calculated in Camera system and the camera intrinsic parameters, I get an image like this:
The target keeps its pattern. But it moved from bottom right to top left. I cannot understand why.
Yes, the camera center is the origin of the camera coordinate system, which seems to be right following to this post.
In case of camera pose estimation, value seems reasonable can be named as backprojection error. That's a measure of how well your resulting rotation and translation map the 3D points to the 2D pixels. Unfortunately, solvePnP does not return a residual error measure. Therefore one has to compute it:
cv::solvePnP(worldPoints, pixelPoints, camIntrinsics, camDistortion, rVec, tVec);
// Use computed solution to project 3D pattern to image
cv::Mat projectedPattern;
cv::projectPoints(worldPoints, rVec, tVec, camIntrinsics, camDistortion, projectedPattern);
// Compute error of each 2D-3D correspondence.
std::vector<float> errors;
for( int i=0; i < corners.size(); ++i)
{
float dx = pixelPoints.at(i).x - projectedPattern.at<float>(i, 0);
float dy = pixelPoints.at(i).y - projectedPattern.at<float>(i, 1);
// Euclidean distance between projected and real measured pixel
float err = sqrt(dx*dx + dy*dy);
errors.push_back(err);
}
// Here, compute max or average of your "errors"
An average backprojection error of a calibrated camera might be in the range of 0 - 2 pixel. According to your two pictures, this would be way more. To me, it looks like a scaling problem. If I am right, you compute the projection yourself. Maybe you can try once cv::projectPoints() and compare.
When it comes to transformations, I learned not to follow my imagination :) The first thing I Do with the returned rVec and tVec is usually creating a 4x4 rigid transformation matrix out of it (I posted once code here). This makes things even less intuitive, but instead it is compact and handy.
Now I know the answers.
Yes, the camera center is the origin of the camera coordinate system.
Consider that the coordinates in the camera system are calculated as (xc,yc,zc). Then xc should be the distance between the camera and
the point in real world in the x direction.
Next, how to determine whether the output matrices are right?
1. as #eidelen points out, backprojection error is one indicative measure.
2. Calculate the coordinates of the points according to their coordinates in the world coordinate system and the matrices.
So why did I get a wrong result(the pattern remained but moved to a different region of the image)?
Parameter cameraMatrix in solvePnP() is a matrix supplying the parameters of the camera's external parameters. In camera matrix, you should use width/2 and height/2 for cx and cy. While I use width and height of the image size. I think that caused the error. After I corrected that and re-calibrated the camera, everything seems fine.
How can I do a basic face alignment on a 2-dimensional image with the assumption that I have the position/coordinates of the mouth and eyes.
Is there any algorithm that I could implement to correct the face alignment on images?
Face (or image) alignment refers to aligning one image (or face in your case) with respect to another (or a reference image/face). It is also referred to as image registration. You can do that using either appearance (intensity-based registration) or key-point locations (feature-based registration). The second category stems from image motion models where one image is considered a displaced version of the other.
In your case the landmark locations (3 points for eyes and nose?) provide a good reference set for straightforward feature-based registration. Assuming you have the location of a set of points in both of the 2D images, x_1 and x_2 you can estimate a similarity transform (rotation, translation, scaling), i.e. a planar 2D transform S that maps x_1 to x_2. You can additionally add reflection to that, though for faces this will most-likely be unnecessary.
Estimation can be done by forming the normal equations and solving a linear least-squares (LS) problem for the x_1 = Sx_2 system using linear regression. For the 5 unknown parameters (2 rotation, 2 translation, 1 scaling) you will need 3 points (2.5 to be precise) for solving 5 equations. Solution to the above LS can be obtained through Direct Linear Transform (e.g. by applying SVD or a matrix pseudo-inverse). For cases of a sufficiently large number of reference points (i.e. automatically detected) a RANSAC-type method for point filtering and uncertainty removal (though this is not your case here).
After estimating S, apply image warping on the second image to get the transformed grid (pixel) coordinates of the entire image 2. The transform will change pixel locations but not their appearance. Unavoidably some of the transformed regions of image 2 will lie outside the grid of image 1, and you can decide on the values for those null locations (e.g. 0, NaN etc.).
For more details: R. Szeliski, "Image Alignment and Stitching: A Tutorial" (Section 4.3 "Geometric Registration")
In OpenCV see: Geometric Image Transformations, e.g. cv::getRotationMatrix2D cv::getAffineTransform and cv::warpAffine. Note though that you should estimate and apply a similarity transform (special case of an affine) in order to preserve angles and shapes.
For the face there is lot of variability in feature points. So it won't be possible to do a perfect fit of all feature points by just affine transforms. The only way to align all the points perfectly is to warp the image given the points. Basically you can do a triangulation of image given the points and do a affine warp of each triangle to get the warped image where all the points are aligned.
Face detection could be handled based on the just eye positions.
Herein, OpenCV, Dlib and MTCNN offers to detect faces and eyes. Besides, it is a python based framework but deepface wraps those methods and offers an out-of-the box detection and alignment function.
detectFace function applies detection and alignment in the background respectively.
#!pip install deepface
from deepface import DeepFace
backends = ['opencv', 'ssd', 'dlib', 'mtcnn']
DeepFace.detectFace("img.jpg", detector_backend = backends[0])
Besides, you can apply detection and alignment manually.
from deepface.commons import functions
img = functions.load_image("img.jpg")
backends = ['opencv', 'ssd', 'dlib', 'mtcnn']
detected_face = functions.detect_face(img = img, detector_backend = backends[3])
plt.imshow(detected_face)
aligned_face = functions.align_face(img = img, detector_backend = backends[3])
plt.imshow(aligned_face)
processed_img = functions.detect_face(img = aligned_face, detector_backend = backends[3])
plt.imshow(processed_img)
There's a section Aligning Face Images in OpenCV's Face Recognition guide:
http://docs.opencv.org/trunk/modules/contrib/doc/facerec/facerec_tutorial.html#aligning-face-images
The script aligns given images at the eyes. It's written in Python, but should be easy to translate to other languages. I know of a C# implementation by Sorin Miron:
http://code.google.com/p/stereo-face-recognition/