A system is proposed for automatic reading of the number of dots on dice in general table game settings. Different from previous dice recognition systems that recognize dice of a specific color using a single top-view camera in an enclosure with controlled settings, the proposed one uses multiple cameras to recognize dice of various colors and under uncontrolled conditions. It is composed of three modules. Module-1 locates the dice using the gradient-conditioned color segmentation, proposed, to segment dice of arbitrary colors from the background. Module-2 exploits the local invariant features good for building homographies, giving a solution to segment the top faces of the dice. To identify the dots on the segmented top faces, a maximally stable extremal region detector is embedded in module-3 for its consistency in locating the dot region. Experiments show that the proposed system performs satisfactorily in various test conditions.