A high-speed stereovision system based on smart cameras is presented to track table tennis ball. A distributed parallel processing architecture is developed to improve the real-time performance of the system. A set of novel target recognition and tracking algorithm is also proposed to distinguish the ball from the changing background with less computation and good robustness. It takes about 6 ms to capture an image and extract the ball. The trajectory of the ball is well measured in real-time. The experimental results are provided to verify the effectiveness of the proposed methods.