Deep Learning-Based Image Kernel for Inductive Transfer
This addresses the challenge of few-shot learning in image classification, which is incremental as it builds on existing transfer learning and Siamese network approaches.
The paper tackles the problem of classifying images from target classes with few training examples by using transfer learning from non-target classes, resulting in a method that outperforms state-of-the-art alternatives when partially fine-tuned.
We propose a method to classify images from target classes with a small number of training examples based on transfer learning from non-target classes. Without using any more information than class labels for samples from non-target classes, we train a Siamese net to estimate the probability of two images to belong to the same class. With some post-processing, output of the Siamese net can be used to form a gram matrix of a Mercer kernel. Coupled with a support vector machine (SVM), such a kernel gave reasonable classification accuracy on target classes without any fine-tuning. When the Siamese net was only partially fine-tuned using a small number of samples from the target classes, the resulting classifier outperformed the state-of-the-art and other alternatives. We share class separation capabilities and insights into the learning process of such a kernel on MNIST, Dogs vs. Cats, and CIFAR-10 datasets.