IDEAL: Independent Domain Embedding Augmentation Learning
This addresses the need for better data transformation techniques in deep metric learning for researchers and practitioners in computer vision, offering an incremental but effective enhancement to existing methods.
The paper tackles the problem of improving deep metric learning by focusing on low-level data transformation, proposing the IDEAL method to learn multiple independent embedding spaces for multiple domains, which enhances performance on visual retrieval tasks, achieving state-of-the-art results with improvements such as 84.5% to 87.1% on Cars-196 and 65.8% to 69.5% on CUB-200 at Recall@1.
Many efforts have been devoted to designing sampling, mining, and weighting strategies in high-level deep metric learning (DML) loss objectives. However, little attention has been paid to low-level but essential data transformation. In this paper, we develop a novel mechanism, the independent domain embedding augmentation learning ({IDEAL}) method. It can simultaneously learn multiple independent embedding spaces for multiple domains generated by predefined data transformations. Our IDEAL is orthogonal to existing DML techniques and can be seamlessly combined with prior DML approaches for enhanced performance. Empirical results on visual retrieval tasks demonstrate the superiority of the proposed method. For example, the IDEAL improves the performance of MS loss by a large margin, 84.5\% $\rightarrow$ 87.1\% on Cars-196, and 65.8\% $\rightarrow$ 69.5\% on CUB-200 at Recall$@1$. Our IDEAL with MS loss also achieves the new state-of-the-art performance on three image retrieval benchmarks, \ie, \emph{Cars-196}, \emph{CUB-200}, and \emph{SOP}. It outperforms the most recent DML approaches, such as Circle loss and XBM, significantly. The source code and pre-trained models of our method will be available at\emph{\url{https://github.com/emdata-ailab/IDEAL}}.