Multi-resolution dictionary learning for face recognition

Highlights：

• In this paper, we propose a novel multi-resolution dictionary learning algorithm for face recognition.

• The proposed method seeks to learn one dictionary for each resolution, ensuring a perfect recovery for each resolution space.

• We propose a joint objective function to capture the patterns from multi-resolution images for face recognition.

• It enforces the similarity of representations obtained using multiple dictionaries in the training phase to enhance the robustness of the model.