Multi-Level Cascade Sparse Representation Learning for Small Data Classification

Deep learning (DL) methods have recently captured much attention for image classification. However, such methods may lead to a suboptimal solution for small-scale data since the lack of training samples. Sparse representation stands out with its efficiency and interpretability, but its precision is not so competitive. We develop a Multi-Level Cascade Sparse Representation (ML-CSR) learning method to combine both advantages when processing small-scale data. ML-CSR is proposed using a pyramid structure to expand the training data size. It adopts two core modules, the Error-To-Feature (ETF) module, and the Generate-Adaptive-Weight (GAW) module, to further improve the precision. ML-CSR calculates the inter-layer differences by the ETF module to increase the diversity of samples and obtains adaptive weights based on the layer accuracy in the GAW module. This helps ML-CSR learn more discriminative features. State-of-the-art results on the benchmark face databases validate the effectiveness of the proposed ML-CSR. Ablation experiments demonstrate that the proposed pyramid structure, ETF, and GAW module can improve the performance of ML-CSR. The code is available at https://github.com/Zhongwenyuan98/ML-CSR.

Automated summary

Deep learning methods have attracted much attention for image classification recently. However, for small-scale data, these methods may not yield optimal results due to the lack of training samples. Sparse representation is efficient and interpretable, but its precision is not competitive. To address this issue, we propose a Multi-Level Cascade Sparse Representation (ML-CSR) learning method that combines the advantages of both deep learning and sparse representation. ML-CSR utilizes a pyramid structure and two core modules, Error-To-Feature (ETF) and Generate-Adaptive-Weight (GAW), to improve precision. Experiments on face databases demonstrate the effectiveness of ML-CSR, and ablation experiments further confirm the benefits of the proposed pyramid structure, ETF, and GAW modules. The code is available at https://github.com/Zhongwenyuan98/ML-CSR.