The Group Loss for Deep Metric Learning
Authors:
Preprint
Abstract
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 show state-of-the-art results on clustering and image retrieval on several datasets, and show the potential of our method when combined with other techniques such as ensemblesKeywords
All Related Versions
This publication has 6 references indexed in Scilit:
- Deep Metric Learning with BIER: Boosting Independent Embeddings RobustlyIEEE Transactions on Pattern Analysis and Machine Intelligence, 2018
- Graph Transduction as a Noncooperative GameNeural Computation, 2012
- Product Quantization for Nearest Neighbor SearchIEEE Transactions on Pattern Analysis and Machine Intelligence, 2010
- The Dynamics of Nonlinear Relaxation Labeling ProcessesJournal of Mathematical Imaging and Vision, 1997
- Scene Labeling by Relaxation OperationsIEEE Transactions on Systems, Man, and Cybernetics, 1976
- VII. Note on regression and inheritance in the case of two parentsProceedings of the Royal Society of London, 1895