In human-made scenarios, robots need to be able to fully operate objects in\ntheir surroundings, i.e., objects are required to be functionally grasped\nrather than only picked. This imposes very strict constraints on the object\npose such that a direct grasp can be performed. Inspired by the anthropomorphic\nnature of humanoid robots, we propose an approach that first grasps an object\nwith one hand, obtaining full control over its pose, and performs the\nfunctional grasp with the second hand subsequently. Thus, we develop a fully\nautonomous pipeline for dual-arm functional regrasping of novel familiar\nobjects, i.e., objects never seen before that belong to a known object\ncategory, e.g., spray bottles. This process involves semantic segmentation,\nobject pose estimation, non-rigid mesh registration, grasp sampling, handover\npose generation and in-hand pose refinement. The latter is used to compensate\nfor the unpredictable object movement during the first grasp. The approach is\napplied to a human-like upper body. To the best knowledge of the authors, this\nis the first system that exhibits autonomous bimanual functional regrasping\ncapabilities. We demonstrate that our system yields reliable success rates and\ncan be applied on-line to real-world tasks using only one off-the-shelf RGB-D\nsensor.\n