AI-powered
podcast player
Listen to all your favourite podcasts with AI-powered features
How Do You Train a 3D Model?
For each mold, we would rotate our orientation, I think about 200 times. So for a random shape in the domain of the deformation and resolution of our 3D printer shapes, we're able to actually see any shape and have a calibrated depth output. We do this 30,000 times. And so that's really how we thought about it. What was the range of depth that we wanted to experience? And how do we ensure that we covered the full area of these sensors?"