Sparse representaion via ℓ1/2-norm minimization for facial expression recognition

The ℓ_1/2-norm regularizer is shown to have many promising properties such as unbiasedness, sparsity and oracle properties. By exploiting these properties of ℓ_1/2-norm regularizer, we propose a novel model of sparse representation based classification via ℓ_1/2-norm minimization (ℓ_1/2-SRC) for facial expression recognition in this paper. In ℓ_1/2-SRC, we use ℓ_1/2-norm minimization as an alternative to ℓ₀-norm minimization instead of using ℓ₁-norm minimization in the traditional ℓ¹-SRC. By adopting ℓ_1/2-norm minimization, we can find a sparser and more accurate solution than the ℓ¹-SRC, and the optimization problem of ℓ_1/2-norm minimization is much easier to be solved than that of ℓ₀-norm minimization. Furthermore, an active-set based iterative reweighted algorithm is proposed to solve the ℓ_1/2-norm minimization problem. The experimental results on JAFFE and Cohn-Kanade databases testify the efficiency of ℓ_1/2-SRC.