Deep Randomized Ensembles for Image Retrieval Open Access
Learning embedding functions, which map semantically related images to nearby locations in a feature space supports a variety of image 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 retrieval performance. 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 the state of the art performance for image retrieval on various retrieval tasks: Birds species retrieval, Similar cars image searching, Fashion clothes matching, etc. Code is available at: https://github.com/littleredxh/DREML
Notice to Authors
If you are the author of this work and you have any questions about the information on this page, please use the Contact form to get in touch with us.