Symmetrical Synthesis For Deep Metric Learning

Abstract

Deep metric learning aims to learn embeddings that contain semantic similarity information among data points. To learn better embeddings, methods to generate synthetic hard samples have been proposed. Existing methods of synthetic hard sample generation are adopting autoencoders or generative adversarial networks, but this leads to more hyper-parameters, harder optimization, and slower training speed. In this paper, we address these problems by proposing a novel method of synthetic hard sample generation called symmetrical synthesis. Given two original feature points from the same class, the proposed method firstly generates synthetic points with each other as an axis of symmetry. Secondly, it performs hard negative pair mining within the original and synthetic points to select a more informative negative pair for computing the metric learning loss. Our proposed method is hyper-parameter free and plug-and-play for existing metric learning losses without network modification. We demonst

Stats

• citations13
• S2 citations—
• github stars0
• HF likes0
• heat score8.60
• arxiv keygu2020symmetrical

Related papers

• Proxy Synthesis: Learning With Synthetic Classes For Deep Metric Learning (2021)10.85
• Dynamic Sampling For Deep Metric Learning (2020)5.84
• Learning To Generate Novel Classes For Deep Metric Learning (2022)0.00
• Loop: Looking For Optimal Hard Negative Embeddings For Deep Metric Learning (2021)8.82
• Embedding Expansion: Augmentation In Embedding Space For Deep Metric Learning (2020)12.02
• The General Pair-based Weighting Loss For Deep Metric Learning (2019)0.00
• Deep Metric Learning Assisted By Intra-variance In A Semi-supervised View Of Learning (2023)5.24
• Sharing Matters For Generalization In Deep Metric Learning (2020)8.35