Affinity guided Geometric Semi-Supervised Metric Learning

2020/02/27 by Ujjal Kr Dutta, Mehrtash Harandi, Dutta, Ujjal Kr +3
Computer Science · #Computer Vision and Pattern Recognition (cs.CV) #FOS: Computer and information sciences #Face and Expression Recognition #Human Pose and Action Recognition #Machine Learning (cs.LG) #Video Surveillance and Tracking Methods

paper · pdf · doi:10.48550/arxiv.2002.12394

openalex publication_date 2020/02/27 · openalex created_date 2025/10/10 · openalex updated_date 2026/07/28

Abstract

In this paper, we revamp the forgotten classical Semi-Supervised Distance Metric Learning (SSDML) problem from a Riemannian geometric lens, to leverage stochastic optimization within a end-to-end deep framework. The motivation comes from the fact that apart from a few classical SSDML approaches learning a linear Mahalanobis metric, deep SSDML has not been studied. We first extend existing SSDML methods to their deep counterparts and then propose a new method to overcome their limitations. Due to the nature of constraints on our metric parameters, we leverage Riemannian optimization. Our deep SSDML method with a novel affinity propagation based triplet mining strategy outperforms its competitors.

Citations

• Gradient-based learning applied to document recognition
• ImageNet Large Scale Visual Recognition Challenge
• Discriminative Unsupervised Feature Learning with Exemplar Convolutional Neural Networks
• Fashion-MNIST: a Novel Image Dataset for Benchmarking Machine Learning Algorithms
• Learning Multiple Layers of Features from Tiny Images

Related

• Semi-Supervised Metric Learning: A Deep Resurrection
• Robust Geometric Metric Learning
• SERAPH: Semi-supervised Metric Learning Paradigm with Hyper Sparsity
• Spectral, Probabilistic, and Deep Metric Learning: Tutorial and Survey
• An Efficient Dual Approach to Distance Metric Learning