MATLAB Answers

What does it mean when I get this error?

2 views (last 30 days)
Christian Conlon
Christian Conlon on 5 Dec 2019
I have been in the process of trying to calibrate two videos taken with two cameras in parallax view using a 0.935m calibration wand.
My steps have been:
1) Cut 2 synced videos down to 10 frames of the same legnth in Kinovea
2) Load DLTdv5 and intitalize both videos (L Camera and R camera).
3) Digitize 4 points per frame for the 10 frames for 40 total points:
-2 points on a light located on either end of of the wand in the L video, and two points locataed on either end of the R video
4) Digitize 20 background points using the same point in each camera with 2 points/frame for ten frames
5) Load easyWand5
6) Load my 40 calibration points
-I get the following message:
Success - 0 wand points shared across 40 cameras were loaded
7) Load my background points
-I recieve the same error message as before
It should be noted that I am using the .csv files MATLAB made when digitizing in steps 3 + 4 for x,y points. The x,y,z .csv files don't appear to have and data in them, just NaN values
8) Enter the wand length 0.935
9) Enter the Image Width, Image Height and calculated FL Estimate in easyWand5
-as mentioned before, easyWand5 is loading my points as 40 separate cameras so I have to do this step for 40 cameras
-I assume I am doing something incorrectly when I am digitizing my intitial callibration file
10) Compute callibraition:
Here I get the following error and this is what I am asking about:
Error using svd
Input to SVD must not contain NaN or Inf.
Error in easyWand5>twoCamCal_v2 (line 2783)
[U,D,V] = svd(E);
Error in easyWand5>sbaCalib_pass2 (line 3915)
[R(:,:,i),tv(:,:,i)] = twoCamCal_v2([ptNorm(:,pCams(i)*2-1:pCams(i)*2), ...
Error in easyWand5>computeCalibrationButton_callback (line 1843)
Error while evaluating UIControl Callback

  1 Comment

Christine Tobler
Christine Tobler on 5 Dec 2019
The SVD function does not accept inputs that contain NaN values. Even if it did, it would only return more NaN values, so your information has been lost way before this step, at the point where you note that your files contain only NaN values.
You could try inspecting your data in the debugger, to see at which point it starts containing NaN values - this would be the operation that's causing the issue.

Sign in to comment.

Answers (0)

Sign in to answer this question.