Show Posts

This section allows you to view all posts made by this member. Note that you can only see posts made in areas you currently have access to.

Topics - shwetankumar

Pages: [1]
General / Export per camera reprojection error as CSV
« on: February 06, 2018, 07:51:46 PM »

I would like to know if there is a script that can perform a similar function as the export button in the reference tab to export per camera : image name, accuracy (m), error (m) and error (pixel) ? I am trying to export this data after camera optimization has been done.



General / Using calibrated parameters without recomputation
« on: June 12, 2017, 06:21:05 AM »

If we have a number of calibrated images (both intrinsic and extrinsic parameters are available), how can we use agisoft photoscan to create a dense point cloud such that those parameters are used as is and agisoft does not to modify them ? For each image, the known parameters are: camera location (x, y, z), camera rotation (3x3 matrix), focal length in pixel, principal point, distortion parameters.


General / Transform for Projecting Point Cloud onto Orthomosaic
« on: March 29, 2017, 03:38:29 AM »

I recently started generating orthomosaic images using Agisoft and the quality of the results have been extremely good. I am trying to do a small experiment for demonstration purpose. I am tusing geotagged images for performing scaled 3D point cloud generation, and following up with orthomosaic generation.
Since the point cloud quality of the model at High density resembles the orthomosaic view very closely, the idea for the demo will be to show how I can project the points of the dense cloud onto the orthomosaic in a way that they line up correctly.
I am guessing there is a transform that can take me from the scaled point cloud (in metres) onto the 2D orthomosaic photo. The transform seems to be an orthographic projection but I am not able to compute the exact projection vector along which I should get consistently good projections. I have tried some random vectors based on hit and trial values, but obtaining the actual transform will be very helpful.
The goal is to show how well the projected point cloud lines up with the orthomosaic. Furthermore, I am trying to show dimensions in each and show that the dimensions in the 2D orthomosaic is the same as the scaled point cloud.



General / Camera Centers not correct in certain projects
« on: March 14, 2017, 05:58:54 PM »

I ran a project through the Photoscan Professional and observed the following :

After successful dense point cloud generation, I exported the camera properties and got the standard xml file (attachment).
Each aligned image has a transformation object in the xml file.
1) I constructed a 4x4 matrix from the values (A).
2) Then I took the inverse of the above 4x4 matrix (A^-1).
3) The upper left 3x3 matrix of A^-1 was taken to be the rotation matrix R.
4) In the 4th column, the top 3 values were taken to be the translation column matrix t.
5) Using C = -Rt, I got the camera center C.
6) I plotted the 3D camera centers along with the point cloud and got the following model (attachment).

Can you help me figure out where I am going wrong ? The camera centers seem to be below the 3D structure.
Also attached is the view inside the Agisoft product.




I am curious to know what lens distortion formula to use for computing projection, given
1) Radial distortion co-efficients k1, k2, k3, k4
2) Tangential distortion co-efficients p1, p2, p3, p4

I found a formula in the Agisoft Lens User Manual Version 0.4.0, but it does not incorporate all the 8 distortion coefficients seen in an exported XML file.

x = X / Z
y = Y / Z

x' = x(1 + K1r2 + K2r4 + K3r6) + P2(r2+2x2) + 2P1xy
y' = y(1 + K1r2 + K2r4 + K3r6) + P1(r2+2y2) + 2P2xy
u = cx + x'fx + y'skew
v = cy + y'fy

Thanks !

Pages: [1]