ChArUco2: Enhanced calibration boards with dual black-and-white marker detection

Camera calibration patterns are fundamental tools in computer vision, robotics, and augmented reality. In recent years, hybrid patterns such as ChArUco, which combine fiducial markers with a traditional checkerboard, have become increasingly common due to their versatility. In particular, the ChArUco board, distributed as part of OpenCV, places markers only in the white squares of a checkerboard, limiting marker density and leaving all black squares unused. The associated ChArUco diamond target is the smallest self-contained configuration usable as a standalone marker. However, it provides only four reference points, restricting pose estimation accuracy. We present ChArUco2 , a C + + library that addresses both limitations. ChArUco2 places a marker in every square: black squares contain standard markers and white squares contain inverted markers. This doubles marker density and allows each marker to occupy the full square area, improving detection range and robustness. The ChArUco2 diamond is a 2 × 2 four-marker block that yields nine reference points spanning the full marker extent, offering a larger geometric baseline for single-shot pose estimation over the standard four-corner diamond. Experimental evaluation under partial occlusion shows that ChArUco2 consistently provides more reference points, detecting e.g., 36 corners compared to 16 for standard ChArUco at 50% occlusion. The library integrates directly with the OpenCV API and supports any ArUco dictionary.