Diversified Mutual Learning for Deep Metric Learning

Mutual learning is an ensemble training strategy to improve generalization by transferring individual knowledge to each other while simultaneously training multiple models . The proposed method with a conventional triplet loss achieves the state-of-the-art performance of [email protected] on standard datasets: 69.9 on CUB-200-2011 and 89.1 on CARS-196 . Our method is particularly adequate for inductive transfer learning at the lack of large-scale data, where the embedding model is initialized with a pretrained model and then fine-tuned on a target dataset. Extensive experiments show that our method significantly improves individual models as well as their ensemble. Our method significantly improved individual models… The method is especially adequate for inductionive transfer learn at the lacks of large scale data, with the exception of

Links: PDF - Abstract

Code :

None

Keywords : method - learning - models - individual - scale -

Leave a Reply

Your email address will not be published. Required fields are marked *