One shot contact learning

We present a general method for one-shot learning of dexterous grasps and object placement. We rely on a product of experts formulation. Experts in the product are of two types. The first is a contact model and is a density over the pose of a single hand link relative to the local object surface. The second is the hand configuration model and is a density over the whole hand configuration. Grasp generation for a novel object optimises the product of these two model types. Selection between grasp types is also automatic. The success rate is 84.4% or 77.7% if seven views or a single view of the test object are taken. We also show initial results on object placement