Deep convolutional neural networks have shown to perform well in across many computer vision tasks. However, many such methods require hundreds, if not thousands, of images per class to generalize well to unseen examples. This is restricting when obtaining and labelling larger volumes of data is impractical, such as observing rare objects/events, performing real-time operations, or operating in new environments. My research investigates Few-Shot Learning (FSL) algorithms that aim to learn using limited data. Finding an algorithm capable of learning from only a few samples could reduce the time spent obtaining and labelling datasets, and accelerate the training of deep-learning models.