Metric Learning with HORDE: High-Order Regularizer for Deep Embeddings

Abstract : Learning an effective similarity metric between the image representations is key to the success of the recent advances in visual search tasks (e.g. verification or zero shot learning). Although the metric learning part is well addressed, this metric is usually computed over the average of the extracted deep features. This representation is then trained to be discriminative, however, these deep features tend to be scattered across the feature space. That way, the representations are not robust for the feature outliers, object occlusions, background variations, etc. In this paper, we tackle this scattering problem with a distribution-aware regularization named HORDE. This regularizer is designed to ensure that the feature distributions from two dissimilar images are well separated in the feature space. In the case of similar images, the feature distributions are enforced to be localized in the same feature space. We prove theoretically that our regularizer upper and lower bounds wellknown distances between distributions, such as the Maximum Mean Discrepancy and the Wasserstein distance. Empirically, HORDE consistently improves deep metric learning architectures, leading to state-of-the-art results on 3 out of 4 deep metric learning datasets.
Document type :
Conference papers
Complete list of metadatas
Contributor : Aymeric Histace <>
Submitted on : Wednesday, October 9, 2019 - 8:56:36 AM
Last modification on : Thursday, October 10, 2019 - 1:29:31 AM

Links full text


  • HAL Id : hal-02309147, version 1
  • ARXIV : 1908.02735


Pierre Jacob, David Picard, Aymeric Histace, Edouard Klein. Metric Learning with HORDE: High-Order Regularizer for Deep Embeddings. International Conference on Computer Vision, Oct 2019, Seoul, South Korea. ⟨hal-02309147⟩



Record views