Learning Intra-Batch Connections for Deep Metric Learning
This addresses the limitation of existing methods that only use pairwise or triplet relations, improving performance for tasks like image retrieval, though it is incremental as it builds on prior metric learning approaches.
The paper tackles the problem of deep metric learning by proposing a message passing network that considers all relations within a mini-batch, using attention to weigh neighbor importance, achieving state-of-the-art results on clustering and image retrieval across datasets like CUB-200-2011 and Cars196.
The goal of metric learning is to learn a function that maps samples to a lower-dimensional space where similar samples lie closer than dissimilar ones. Particularly, deep metric learning utilizes neural networks to learn such a mapping. Most approaches rely on losses that only take the relations between pairs or triplets of samples into account, which either belong to the same class or two different classes. However, these methods do not explore the embedding space in its entirety. To this end, we propose an approach based on message passing networks that takes all the relations in a mini-batch into account. We refine embedding vectors by exchanging messages among all samples in a given batch allowing the training process to be aware of its overall structure. Since not all samples are equally important to predict a decision boundary, we use an attention mechanism during message passing to allow samples to weigh the importance of each neighbor accordingly. We achieve state-of-the-art results on clustering and image retrieval on the CUB-200-2011, Cars196, Stanford Online Products, and In-Shop Clothes datasets. To facilitate further research, we make available the code and the models at https://github.com/dvl-tum/intra_batch_connections.