Currently the test of the triangulator position is based on a single assertion about mean point positions, but this is imprecise and I have discovered shifts slightly as underlying packages are updated. I'd like frame-by-frame assessment of actual 3D distance.
Currently the test of the triangulator position is based on a single assertion about mean point positions, but this is imprecise and I have discovered shifts slightly as underlying packages are updated. I'd like frame-by-frame assessment of actual 3D distance.