Abstract: Methods, systems, apparatus, and computer-readable media (transitory or non-transitory) are described herein for scaling and rendering a virtual hand. According to an example, vision data may be received from a three-dimensional (“3D”) vision sensor. The vision data may capture at least a portion of a user in an environment, and may include data representing the user's hand relative to a touch interaction surface. The vision data may be processed to generate a 3D representation of the user's hand. A scaling center may be identified on the touch interaction surface to scale the 3D representation of the user's hand. The 3D representation of the user's hand may be scaled with respect to the identified scaling center using a scaling factor. The scaling factor may be based on a rendering constraint. A virtual hand may be rendered, e.g., on a display, based on the scaled 3D representation of the user's hand.