Skip to main content

Learning local descriptors with a CDF-based dynamic soft margin

Author(s): Zhang, L; Rusinkiewicz, Szymon

To refer to this page use:
Abstract: © 2019 IEEE. The triplet loss is adopted by a variety of learning tasks, such as local feature descriptor learning. However, its standard formulation with a hard margin only leverages part of the training data in each mini-batch. Moreover, the margin is often empirically chosen or determined through computationally expensive validation, and stays unchanged during the entire training session. In this work, we propose a simple yet effective method to overcome the above limitations. The core idea is to replace the hard margin with a non-parametric soft margin, which is dynamically updated. The major observation is that the difficulty of a triplet can be inferred from the cumulative distribution function of the triplets' signed distances to the decision boundary. We demonstrate through experiments on both real-valued and binary local feature descriptors that our method leads to state-of-the-art performance on popular benchmarks, while eliminating the need to determine the best margin.
Publication Date: 1-Oct-2019
Citation: Zhang, L, Rusinkiewicz, S. (2019). Learning local descriptors with a CDF-based dynamic soft margin. Proceedings of the IEEE International Conference on Computer Vision, 2019-October (2969 - 2978. doi:10.1109/ICCV.2019.00306
DOI: doi:10.1109/ICCV.2019.00306
ISSN: 1550-5499
Pages: 2969 - 2978
Type of Material: Journal Article
Journal/Proceeding Title: Proceedings of the IEEE International Conference on Computer Vision
Version: Final published version. This is an open access article.

Items in OAR@Princeton are protected by copyright, with all rights reserved, unless otherwise indicated.