Metric Learning With HORDE: High-order Regularizer For Deep Embeddings

Abstract

Learning an effective similarity measure between image representations is key to the success of 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. Consequently, the representations are not robust to outliers, object occlusions, background variations, etc. In this paper, we tackle this scattering problem with a distribution-aware regularization named HORDE. This regularizer enforces visually-close images to have deep features with the same distribution which are well localized in the feature space. We provide a theoretical analysis supporting this regularization effect. We also show the effectiveness of our approach by obtaining state-of-the-art results on 4 well-known datasets (Cub-200-2011,

Stats

• citations50
• S2 citations—
• github stars0
• HF likes0
• heat score12.81
• arxiv keyjacob2019metric

Related papers

• Deep Metric Learning Using Similarities From Nonlinear Rank Approximations (2019)2.26
• Learning Deep Optimal Embeddings With Sinkhorn Divergences (2022)0.00
• Loop: Looking For Optimal Hard Negative Embeddings For Deep Metric Learning (2021)8.82
• Hyp\(^2\) Loss: Beyond Hypersphere Metric Space For Multi-label Image Retrieval (2022)13.32
• Multi-level Distance Regularization For Deep Metric Learning (2021)8.09
• Directional Statistics-based Deep Metric Learning For Image Classification And Retrieval (2018)13.05
• Hierarchical Metric Learning And Matching For 2D And 3D Geometric Correspondences (2018)11.39
• Local Similarity-aware Deep Feature Embedding (2016)0.00