Thanks Erik!
We'd be looking at getting ~ 5 units for 15 cameras total. What do you expect the price to be on that?
Wouldn't need the internal stereo depth feature since we'd be feeding data into a NeRF algorithm (which might mean that using these cameras is somewhat unsuitable, but I thought that the 'official OpenCV camera' is a good thing to include in the candidates).
Elliot
Thank you
Elliot
Just to check - by any chance did you open and then reseal the device in a humid warm environment?
(and then put the device outdoors in a cooler environment where the warm humid air inside would condense)
Generally when something is sealed shut with an o-ring (like this device), humidity can not enter/exit the device either in liquid form or in air. If it enters as a liquid then there should be a tell-tale sign where the leak occured (e.g. water pools near to where the leak occured / there's a 'wetter area' inside near to where the liquid entered).
If water entered by condensing humid air, then you'd see the humidity evenly spread across all the components (which appears to be the case here), which might mean that it has been previously sealed in a warm, humid environment.
However, it's very difficult to say for sure. You could try:
I know this isn't useful if all you want is a working device and all you have right now is a wet one 
. But it might help for testing the cause of the issue.
Hey all!
Great to see the OAK develop.
I'm wondering if it's possible to configure the OAK-D Pro with 3 * IMX379 wide angle cameras?
It seems elsewhere that it is mentioned that in general you can switch the cameras between IMX379 and OV9282.
Looking to make an outdoors NeRF capture rig for a project with many color cameras working outdoors, and the OAK-D Pro PoE IP67 system is a great baseline.
Elliot
