Alert button

Hardness-Aware Deep Metric Learning

Add code
Alert button
Mar 13, 2019
Figure 1 for Hardness-Aware Deep Metric Learning
Figure 2 for Hardness-Aware Deep Metric Learning
Figure 3 for Hardness-Aware Deep Metric Learning
Figure 4 for Hardness-Aware Deep Metric Learning

Share this with someone who'll enjoy it:

This paper presents a hardness-aware deep metric learning (HDML) framework. Most previous deep metric learning methods employ the hard negative mining strategy to alleviate the lack of informative samples for training. However, this mining strategy only utilizes a subset of training data, which may not be enough to characterize the global geometry of the embedding space comprehensively. To address this problem, we perform linear interpolation on embeddings to adaptively manipulate their hard levels and generate corresponding label-preserving synthetics for recycled training, so that information buried in all samples can be fully exploited and the metric is always challenged with proper difficulty. Our method achieves very competitive performance on the widely used CUB-200-2011, Cars196, and Stanford Online Products datasets.

* Accepted as CVPR 2019 Oral. Source code available at https://github.com/wzzheng/HDML  

Share this with someone who'll enjoy it: