Meshmapping for demi-spherical dome projection

[Bram Nuttin]

Hi everyone,

I am trying to setup bonsai for visual stimulation in a demi-spherical dome using the bonvision package. The visual scene would be projected on the inside of the dome using a fisheye lens.

I found the calibration steps for a normal screen on https://github.com/bonvision/examples/tree/master/ScreenCalibration, but I think I need to add a MeshMapping node when displaying the checkerboard and aruco images. The meshmapping node requires a csv file to be provided, but how is this csv file constructed? I found that it requires the coordinates of the intersections between meridians and parallels in degrees and pixel units, and normalised brightness levels. Do I manually enter the coordinates and pixel units in the csv file, or is there an automated way of doing this?

Thanks in advance!

Bram

[eabho...@gmail.com]

Hi Bram,

We have created some Bonsai workflows as well as a walk-through guide to do exactly this. The plan is to upload it tomorrow - I'll reply again once its up on the website.

Edd

[Bram]

That is great to hear! Thanks for letting me know.

Bram

[eabho...@gmail.com]

The guide can be found here, along with a link to the relevant Bonsai workflows. It's still under construction for now.
 
https://bonvision.github.io/demos/demos_006_MeshMapping/

Let us know if you have any questions or issues.

Edd

[Bram]

Thank you for uploading the documentation! I have tried to do the calibration, but have encountered an issue. 

After doing the calibration, a strong warping remains when presenting a circle in the upper visual field (I added an image of this in the attachments). Should I execute another step before presenting stimuli on the screen?

Also, in the refinement step I noticed that the nodes were flipped vertically so that the text boxes indicated lower visual angles for higher nodes and vice versa. I therefore repeated step 1 of the calibration, but I worked from low to high visual angles. Could this, and perhaps also the warping issue, be caused by the position of the projector at the back of the dome? The text on the meshmap image is also mirrored and upside down.

I have added the meshmapping output, my stimulus bonsai code, and an image of the calibrated grid in the attachments.

[Bram]

Here is what a circle looks like in the upper visual field.