The pose changes involved in face images have always been a difficulty in face recognition. In order to solve this difficulty, we propose a pose face recognition method based on supervised subspace learning (SSL). In SSL, we first propose a supervised subspace learning algorithm (SSLA). On the one hand, SSLA reduces the intra-class differences by constructing the difference term. On the other hand, the samples can be represented by the samples of the same category by constructing the block-diagonal regularization term. Besides, SSLA is not disturbed by noise by constructing a noise robust term. Then, the samples in the original space are mapped to the learned subspace by using SSLA. Finally, those mapped samples are classified by ESRC. The experimental results on Georgia Tech dataset and FEI dataset show that SSL has achieved good recognition results.