Abstract: Canonical correlation analysis(CCA) is an unsupervised subspace learning algorithm. The discriminative power of extracted CCA features can be improved by introducing label information of samples in recognition. Borrowing the idea of relative attributes, we propose a relative strength based supervised CCA algorithm to effectively exploit the label information of multi-view data. The proposed algorithm not only makes the correlation between two-set canonical projections maximum, but also their every component capable of describing the relative strength between different samples. The extracted canonical features by our algorithm are very discriminative in classification due to the use of the strategy that the difference between inter-class samples is larger than that between intra-class samples. Many experimental results on handwritten digit, face and object image datasets show that the proposed algorithm has better performance than existing canonical correlation feature extraction algorithms.