Revisiting Training Strategies and Generalization Performance in Deep Metric Learning
Deep Metric Learning (DML) is arguably one of the most influential lines of research for learning visual similarities with many proposed approaches every year. Although the field benefits from the rapid progress, the divergence in training protocols, architectures, and parameter choices make an unbi...
Main Authors: | , , , , , |
---|---|
Format: | Article in Journal/Newspaper |
Language: | unknown |
Published: |
arXiv
2020
|
Subjects: | |
Online Access: | https://dx.doi.org/10.48550/arxiv.2002.08473 https://arxiv.org/abs/2002.08473 |
id |
ftdatacite:10.48550/arxiv.2002.08473 |
---|---|
record_format |
openpolar |
spelling |
ftdatacite:10.48550/arxiv.2002.08473 2023-05-15T16:01:12+02:00 Revisiting Training Strategies and Generalization Performance in Deep Metric Learning Roth, Karsten Milbich, Timo Sinha, Samarth Gupta, Prateek Ommer, Björn Cohen, Joseph Paul 2020 https://dx.doi.org/10.48550/arxiv.2002.08473 https://arxiv.org/abs/2002.08473 unknown arXiv arXiv.org perpetual, non-exclusive license http://arxiv.org/licenses/nonexclusive-distrib/1.0/ Computer Vision and Pattern Recognition cs.CV FOS Computer and information sciences Article CreativeWork article Preprint 2020 ftdatacite https://doi.org/10.48550/arxiv.2002.08473 2022-03-10T16:04:11Z Deep Metric Learning (DML) is arguably one of the most influential lines of research for learning visual similarities with many proposed approaches every year. Although the field benefits from the rapid progress, the divergence in training protocols, architectures, and parameter choices make an unbiased comparison difficult. To provide a consistent reference point, we revisit the most widely used DML objective functions and conduct a study of the crucial parameter choices as well as the commonly neglected mini-batch sampling process. Under consistent comparison, DML objectives show much higher saturation than indicated by literature. Further based on our analysis, we uncover a correlation between the embedding space density and compression to the generalization performance of DML models. Exploiting these insights, we propose a simple, yet effective, training regularization to reliably boost the performance of ranking-based DML models on various standard benchmark datasets. Code and a publicly accessible WandB-repo are available at https://github.com/Confusezius/Revisiting_Deep_Metric_Learning_PyTorch. : ICML 2020. Main paper 8.25 pages, 26 pages total Article in Journal/Newspaper DML DataCite Metadata Store (German National Library of Science and Technology) |
institution |
Open Polar |
collection |
DataCite Metadata Store (German National Library of Science and Technology) |
op_collection_id |
ftdatacite |
language |
unknown |
topic |
Computer Vision and Pattern Recognition cs.CV FOS Computer and information sciences |
spellingShingle |
Computer Vision and Pattern Recognition cs.CV FOS Computer and information sciences Roth, Karsten Milbich, Timo Sinha, Samarth Gupta, Prateek Ommer, Björn Cohen, Joseph Paul Revisiting Training Strategies and Generalization Performance in Deep Metric Learning |
topic_facet |
Computer Vision and Pattern Recognition cs.CV FOS Computer and information sciences |
description |
Deep Metric Learning (DML) is arguably one of the most influential lines of research for learning visual similarities with many proposed approaches every year. Although the field benefits from the rapid progress, the divergence in training protocols, architectures, and parameter choices make an unbiased comparison difficult. To provide a consistent reference point, we revisit the most widely used DML objective functions and conduct a study of the crucial parameter choices as well as the commonly neglected mini-batch sampling process. Under consistent comparison, DML objectives show much higher saturation than indicated by literature. Further based on our analysis, we uncover a correlation between the embedding space density and compression to the generalization performance of DML models. Exploiting these insights, we propose a simple, yet effective, training regularization to reliably boost the performance of ranking-based DML models on various standard benchmark datasets. Code and a publicly accessible WandB-repo are available at https://github.com/Confusezius/Revisiting_Deep_Metric_Learning_PyTorch. : ICML 2020. Main paper 8.25 pages, 26 pages total |
format |
Article in Journal/Newspaper |
author |
Roth, Karsten Milbich, Timo Sinha, Samarth Gupta, Prateek Ommer, Björn Cohen, Joseph Paul |
author_facet |
Roth, Karsten Milbich, Timo Sinha, Samarth Gupta, Prateek Ommer, Björn Cohen, Joseph Paul |
author_sort |
Roth, Karsten |
title |
Revisiting Training Strategies and Generalization Performance in Deep Metric Learning |
title_short |
Revisiting Training Strategies and Generalization Performance in Deep Metric Learning |
title_full |
Revisiting Training Strategies and Generalization Performance in Deep Metric Learning |
title_fullStr |
Revisiting Training Strategies and Generalization Performance in Deep Metric Learning |
title_full_unstemmed |
Revisiting Training Strategies and Generalization Performance in Deep Metric Learning |
title_sort |
revisiting training strategies and generalization performance in deep metric learning |
publisher |
arXiv |
publishDate |
2020 |
url |
https://dx.doi.org/10.48550/arxiv.2002.08473 https://arxiv.org/abs/2002.08473 |
genre |
DML |
genre_facet |
DML |
op_rights |
arXiv.org perpetual, non-exclusive license http://arxiv.org/licenses/nonexclusive-distrib/1.0/ |
op_doi |
https://doi.org/10.48550/arxiv.2002.08473 |
_version_ |
1766397164126208000 |