Depth generation is a key technology in computer vision. Known methods mainly rely on stereo matching, or measuring devices like ToF camera. The ToF camera performs well in low-textured regions and repetitive regions where stereo matching fails. In contrast, stereo matching works better than ToF camera in textured regions. Based on their complementary characteristics, we introduce a method to combine ToF depth and stereo matching. In order to integrate their respective advantages, we measure their reliabilities and construct a new cost volume. Experiment results show that our fusion algorithm improves the accuracy and robustness, the generated depth map is much better than that obtained from an individual method.