We present and characterize a simple method for detecting pointing gestures\nsuitable for human-robot interaction applications using a commodity RGB-D\ncamera. We exploit a state-of-the-art Deep CNN-based detector to find hands and\nfaces in RGB images, then examine the corresponding depth channel pixels to\nobtain full 3D pointing vectors. We test several methods of estimating the hand\nend-point of the pointing vector. The system runs at better than 30Hz on\ncommodity hardware: exceeding the frame rate of typical RGB-D sensors. An\nestimate of the absolute pointing accuracy is found empirically by comparison\nwith ground-truth data from a VICON motion-capture system, and the useful\ninteraction volume established. Finally, we show an end-to-end test where a\nrobot estimates where the pointing vector intersects the ground plane, and\nreport the accuracy obtained. We provide source code as a ROS node, with the\nintention of contributing a commodity implementation of this common component\nin HRI systems.\n