RGB color spaces are widely used in image processing. In this paper the authors discuss the use of RGB color space in which the origin is centered in the RGB cube at middle gray. This may appear a trivial difference from the traditional RGB color space in which the origin is at black, yet it is shown that significant advantages accrue from off-setting the origin, as is traditionally done in signal processing with bipolar samples. The geometry of vectors (pixel values) in gray-centered RGB space is described, and related to familiar properties from the hue-saturation-intensity (HSI) representation, which is explained as a different coordinate system, based on RGB. Finally, to illustrate the geometric advantage of centering the origin, the paper presents a simple but effective clipping algorithm for dealing with out of range vectors which can arise from filtering operations applied in RGB space. The effectiveness of the clipping algorithm is demonstrated with some example images.