The Group Loss++: A deeper look into group loss for deep metric learning
This work addresses the challenge of designing effective loss functions for deep metric learning, which is crucial for tasks like image retrieval and person re-identification, though it appears incremental as it builds upon existing group-based methods.
The authors tackled the problem of improving deep metric learning for clustering and image retrieval by proposing Group Loss++, a differentiable label-propagation loss that enforces embedding similarity within groups and separation between groups, achieving state-of-the-art results on four retrieval datasets and competitive results on two person re-identification datasets.
Deep metric learning has yielded impressive results in tasks such as clustering and image retrieval by leveraging neural networks to obtain highly discriminative feature embeddings, which can be used to group samples into different classes. Much research has been devoted to the design of smart loss functions or data mining strategies for training such networks. Most methods consider only pairs or triplets of samples within a mini-batch to compute the loss function, which is commonly based on the distance between embeddings. We propose Group Loss, a loss function based on a differentiable label-propagation method that enforces embedding similarity across all samples of a group while promoting, at the same time, low-density regions amongst data points belonging to different groups. Guided by the smoothness assumption that "similar objects should belong to the same group", the proposed loss trains the neural network for a classification task, enforcing a consistent labelling amongst samples within a class. We design a set of inference strategies tailored towards our algorithm, named Group Loss++ that further improve the results of our model. We show state-of-the-art results on clustering and image retrieval on four retrieval datasets, and present competitive results on two person re-identification datasets, providing a unified framework for retrieval and re-identification.