Introducing shape priors in Siamese networks for image classification

The efficiency of deep neural networks is increasing, and so is the amount of annotated data required for training them. We propose a solution improving the learning process of a classification network with less labeled data. Our approach is to inform the classifier of the elements it should focus on to make its decision by supplying it with some shape priors. These shape priors are expressed as binary masks, giving a rough idea of the shape of the relevant elements for a given class. We resort to Siamese architecture and feed it with image/mask pairs. By inserting shape priors, only the relevant features are retained. This provides the network with significant generalization power without requiring a specific domain adaptation step. This solution is tested on some standard cross-domain digit classification tasks and on a real-world video surveillance application. Extensive tests show that our approach outperforms the classical classifier by generating a good latent space with less training data. Code is available at https://github.com/halqasir/MG-Siamese.

Automated summary

The article proposes a solution to improve the learning process of a classification network by providing shape priors, reducing the need for annotated data. The solution is tested on cross-domain digit classification tasks and a video surveillance application.