This article presents a technique for determining and visual izing the geometric motion capabilities of dual-arm robotic systems when the arms work on an object in a closed kine matic chain configuration, taking account of robot arms' base placements, object dimensions, object holding and contact constraints, and space occupancy conflicts of the two arms' links. The constrained and object orientation restricted motion space in general can be visualized as a complex 3D object with hidden unreachable holes or cavities of varying shapes. An automated visualization methodology is presented together with its graphical implementation, illustrated by an example. The methodology is an inverse computer vision technique in the sense that it creates rather than recognizes visual forms.