Deep Randomized Ensembles for Metric Learning
This work addresses the need for improved metric learning in image retrieval, offering a generalizable and fast solution that enhances performance across multiple datasets.
The paper tackles the problem of learning embedding functions for classification and retrieval tasks by proposing a method that creates ensembles of embeddings through random bagging of training labels, achieving state-of-the-art performance on image retrieval datasets such as CUB-200-2011, Cars-196, In-Shop Clothes Retrieval, and VehicleID.
Learning embedding functions, which map semantically related inputs to nearby locations in a feature space supports a variety of classification and information retrieval tasks. In this work, we propose a novel, generalizable and fast method to define a family of embedding functions that can be used as an ensemble to give improved results. Each embedding function is learned by randomly bagging the training labels into small subsets. We show experimentally that these embedding ensembles create effective embedding functions. The ensemble output defines a metric space that improves state of the art performance for image retrieval on CUB-200-2011, Cars-196, In-Shop Clothes Retrieval and VehicleID.