Triangulation error

[serabos]

Hello!
I am working with pose estimation videos to predict mouse behaviour.
I ran into an error in Label3D.m file today saying after labelling points:

labelGui.triangulateView()
Error using vision.internal.cameraIntrinsicsImpl>parseInputsSimulation
The value of 'principalPoint' is invalid. Expected principalPoint to be positive.

Error in vision.internal.cameraIntrinsicsImpl>parseInputs (line 191)
r = parseInputsSimulation(defaultParams,varargin{:});

Error in vision.internal.cameraIntrinsicsImpl (line 82)
r = parseInputs(varargin{:});

Error in cameraIntrinsics (line 67)
this = this@vision.internal.cameraIntrinsicsImpl(varargin{:});

Error in vision.internal.calibration.CameraParametersImpl/get.Intrinsics (line 502)
intrinsics = cameraIntrinsics(this.FocalLength, ...

Error in Label3D>@(X)X.Intrinsics (line 525)
intrinsics = cellfun(@(X) X.Intrinsics, obj.cameraParams,'uni',0);

Error in Label3D/triangulateView (line 525)
intrinsics = cellfun(@(X) X.Intrinsics, obj.cameraParams,'uni',0);

---

What can be the reason for this? Is it because of the intrinsic files that are being used, or something else?
I would really apppreciate if you could help me out!
Thanks!

[serabos]

^^^ sorry, not Label3D.m file, it should be example.m file!

[diegoaldarondo]

Hi Sera, Yes, historically this issue has always been a result of problems with the intrinsic calibration.

…

On Fri, Sep 15, 2023 at 11:33 AM Sera Bostan ***@***.***> wrote: Hello! I am working with pose estimation videos to predict mouse behaviour. I ran into an error in Label3D.m file today saying after labelling points: labelGui.triangulateView() Error using vision.internal.cameraIntrinsicsImpl>parseInputsSimulation The value of 'principalPoint' is invalid. Expected principalPoint to be positive. Error in vision.internal.cameraIntrinsicsImpl>parseInputs (line 191) r = parseInputsSimulation(defaultParams,varargin{:}); Error in vision.internal.cameraIntrinsicsImpl (line 82) r = parseInputs(varargin{:}); Error in cameraIntrinsics (line 67) this = ***@***.***(varargin{:}); Error in vision.internal.calibration.CameraParametersImpl/get.Intrinsics (line 502) intrinsics = cameraIntrinsics(this.FocalLength, ... Error in Label3D>@(X)X.Intrinsics (line 525) intrinsics = cellfun(@(X) X.Intrinsics, obj.cameraParams,'uni',0); Error in Label3D/triangulateView (line 525) intrinsics = cellfun(@(X) X.Intrinsics, obj.cameraParams,'uni',0); ------------------------------ What can be the reason for this? Is it because of the intrinsic files that are being used, or something else? I would really apppreciate if you could help me out! Thanks! — Reply to this email directly, view it on GitHub <#22>, or unsubscribe <https://github.com/notifications/unsubscribe-auth/ACYHDXEGCQ5I67BUEPCHAKTX2RYMRANCNFSM6AAAAAA4Z7SFAQ> . You are receiving this because you are subscribed to this thread.Message ID: ***@***.***>

[serabos]

Thank you for your quick reply Diego!
So, is it directly because how checkerboard images have been acquired, meaning do I need to re-record them?
Thanks

[diegoaldarondo]

It could be an issue with the images, the automated point detection, or both. When I've encountered this in the past, redoing the intrinsic calibration for the camera(s) in question has been the most efficient approach. If that's not possible, there are other options, but they depend on exactly how off the intrinsics are to begin with.

…

On Fri, Sep 15, 2023 at 11:51 AM Sera Bostan ***@***.***> wrote: Thank you for your quick reply Diego! So, is it directly because how checkerboard images have been acquired, meaning do I need to re-record them? Thanks — Reply to this email directly, view it on GitHub <#22 (comment)>, or unsubscribe <https://github.com/notifications/unsubscribe-auth/ACYHDXAF3YWQHUS6TPAQRZTX2R2OPANCNFSM6AAAAAA4Z7SFAQ> . You are receiving this because you commented.Message ID: ***@***.***>

[serabos]

After redoing the calibration, still getting the same error ufortunately :(
Is there a way I can check how off the intrinsics are? What would be the reference?
Thanks!

[diegoaldarondo]

The main consideration is the location of the principal point. I would follow the recommendations in this thread https://www.mathworks.com/matlabcentral/answers/1771950-negative-y-value-for-principal-point, particularly the parts about evaluating the automatic point detection, limiting the number of images used, and removing outlier images from the calibration.

…

On Fri, Sep 15, 2023 at 1:19 PM Sera Bostan ***@***.***> wrote: After redoing the calibration, still getting the same error ufortunately :( Is there a way I can check how off the intrinsics are? What would be the reference? Thanks! — Reply to this email directly, view it on GitHub <#22 (comment)>, or unsubscribe <https://github.com/notifications/unsubscribe-auth/ACYHDXGMREM6XVNZVIHBK7DX2SEZ5ANCNFSM6AAAAAA4Z7SFAQ> . You are receiving this because you commented.Message ID: ***@***.***>

[serabos]

Okay, I'll look into this now. Thank you very much for your help!