Attention-based Ensemble for Deep Metric Learning

Abstract

Deep metric learning aims to learn an embedding function, modeled as deep neural network. This embedding function usually puts semantically similar images close while dissimilar images far from each other in the learned embedding space. Recently, ensemble has been applied to deep metric learning to yield state-of-the-art results. As one important aspect of ensemble, the learners should be diverse in their feature embeddings. To this end, we propose an attention-based ensemble, which uses multiple attention masks, so that each learner can attend to different parts of the object. We also propose a divergence loss, which encourages diversity among the learners. The proposed method is applied to the standard benchmarks of deep metric learning and experimental results show that it outperforms the state-of-the-art methods by a significant margin on image retrieval tasks.

arXiv commentECCV 2018 camera-ready

References in corpus (6)

• Batch Normalization: Accelerating Deep Network Training by Reducing Internal Covariate Shift
• Recurrent Models of Visual Attention
• Learning Deep Embeddings with Histogram Loss
• Why M Heads are Better than One: Training a Diverse Ensemble of Deep Networks
• No Fuss Distance Metric Learning using Proxies
• Smart Mining for Deep Metric Learning

Cited by in corpus (9)

• Circle Loss: A Unified Perspective of Pair Similarity Optimization
• Deep Metric Learning-based Image Retrieval System for Chest Radiograph and its Clinical Applications in COVID-19
• Fashion Retrieval via Graph Reasoning Networks on a Similarity Pyramid
• Generalization in Metric Learning: Should the Embedding Layer be the Embedding Layer?
• Calibrated neighborhood aware confidence measure for deep metric learning
• Adapting Grad-CAM for Embedding Networks
• Where to Look and How to Describe: Fashion Image Retrieval with an Attentional Heterogeneous Bilinear Network
• Proxy Synthesis: Learning with Synthetic Classes for Deep Metric Learning
• Learning Embeddings for Product Visual Search with Triplet Loss and Online Sampling