Timezone: »

Revisiting Training Strategies and Generalization Performance in Deep Metric Learning
Karsten Roth · Timo Milbich · Samrath Sinha · Prateek Gupta · Bjorn Ommer · Joseph Paul Cohen

Thu Jul 16 12:00 PM -- 12:45 PM & Fri Jul 17 01:00 AM -- 01:45 AM (PDT) @

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/RevisitingDeepMetricLearningPyTorch.

Author Information

Karsten Roth (Heidelberg University, Mila)
Timo Milbich (Heidelberg University)
Samrath Sinha (University of Toronto)
Prateek Gupta (University of Oxford)
Bjorn Ommer (Heidelberg University)
Joseph Paul Cohen (Mila, University of Montreal)

More from the Same Authors