↪︎ ArcFace and Triplet Network
In this project, I implemented a Triplet Network with various online mining strategies to improve training, and the Angular Margin Loss from ArcFace. The goal was to create latent representations of plankton images and compare performance on the downstream task of training a simple classifier on embeddings. Both models achieved good separation on average for all classes in the training data, but struggled with the last class in the unseen data. The ArcFace embeddings showed more connected clusters in the unseen data compared to the Triplet Loss embeddings.
I did this project as part of the course INF368A on representation learning at the University of Bergen during my PhD. The implementation can serve as a starting point for anyone interested in exploring metric learning further or applying it to other tasks.
