Abstract: Shapes and solids modeled by a 3-D computer graphics system have "intelligence", i.e., the shape typically includes a wide variety of characteristics in addition to geometric (e.g., size and positional) information. These additional characteristics permit a user to manipulate 3-D shapes intuitively using a graphical user interface providing "drag and drop" capabilities--reducing the number of factors the user must worry about, saving a lot of time and user frustration, and allowing even non-technical, unskilled users to rapidly and efficiently create impressive images. A component-based architecture may expose any of multiple, alternate representations of the same shape depending on the particular need or application to provide an "intelligent load" that is especially powerful and advantageous in remote interactive imaging applications (e.g., via the Internet) where data transfer and/or load times must be minimized.