Hello,
Thanks for your work. I am more trying to get a solution where I have a single ego centric camera and trying get the pose of the hand of the robot. Your work seems to be using multiple camera is I am not incorrect. Will it be good with single cameras ?
Another question is regarding the training pipeline from synthetic data generation for the articulated robot hand inside the Mb-ICG folder. I don't see any code for that to generate the dataset. Are you planning to open source them ?