This paper is concerned with real-time incremental face recognition for home service robots. For this purpose, we use an Incremental Tensor Subspace Analysis (ITSA) to enroll and recognize the user's face through low-price robot camera when robot performs the real-time customized service in home environments. Furthermore, we perform face recognition based on face database constructed in u-robot test bed environments in ETRI (Electronics and Telecommunications Research Institute). Thus it can be used as a core technique in conjunction with Human-Robot Interaction (HRI) that can naturally interact between human and robots in home robot applications. The experimental results on face database reveal that the presented method shows a good performance in comparison with the well-known methods in distant-varying environments.