Spherical Feature Transform for Deep Metric Learning
This addresses a bottleneck in deep metric learning for computer vision applications, offering an incremental but effective enhancement over previous methods.
The paper tackled the problem of feature augmentation in deep metric learning by proposing a spherical feature transform that uses rotation instead of translation, achieving state-of-the-art results with consistent performance improvements across benchmarks.
Data augmentation in feature space is effective to increase data diversity. Previous methods assume that different classes have the same covariance in their feature distributions. Thus, feature transform between different classes is performed via translation. However, this approach is no longer valid for recent deep metric learning scenarios, where feature normalization is widely adopted and all features lie on a hypersphere. This work proposes a novel spherical feature transform approach. It relaxes the assumption of identical covariance between classes to an assumption of similar covariances of different classes on a hypersphere. Consequently, the feature transform is performed by a rotation that respects the spherical data distributions. We provide a simple and effective training method, and in depth analysis on the relation between the two different transforms. Comprehensive experiments on various deep metric learning benchmarks and different baselines verify that our method achieves consistent performance improvement and state-of-the-art results.