Please use this identifier to cite or link to this item:
https://hdl.handle.net/2440/123378
Citations | ||
Scopus | Web of Science® | Altmetric |
---|---|---|
?
|
?
|
Type: | Conference paper |
Title: | A theoretically sound upper bound on the triplet loss for improving the efficiency of deep distance metric learning |
Author: | Do, T.-T. Tran, T. Reid, I. Kumar, V. Hoang, T. Carneiro, G. |
Citation: | Proceedings / CVPR, IEEE Computer Society Conference on Computer Vision and Pattern Recognition. IEEE Computer Society Conference on Computer Vision and Pattern Recognition, 2019, vol.2019-June, pp.10396-10405 |
Publisher: | IEEE |
Issue Date: | 2019 |
Series/Report no.: | IEEE Conference on Computer Vision and Pattern Recognition |
ISBN: | 9781728132945 |
ISSN: | 1063-6919 2575-7075 |
Conference Name: | IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR) (16 Jun 2019 - 20 Jun 2019 : Long Beach, CA, USA) |
Statement of Responsibility: | Thanh-Toan Do, Toan Tran, Ian Reid, Vijay Kumar, Tuan Hoang, Gustavo Carneiro |
Abstract: | We propose a method that substantially improves the efficiency of deep distance metric learning based on the optimization of the triplet loss function. One epoch of such training process based on a naive optimization of the triplet loss function has a run-time complexity O(N^³), where N is the number of training samples. Such optimization scales poorly, and the most common approach proposed to address this high complexity issue is based on sub-sampling the set of triplets needed for the training process. Another approach explored in the field relies on an ad-hoc linearization (in terms of N) of the triplet loss that introduces class centroids, which must be optimized using the whole training set for each mini-batch - this means that a na"ive implementation of this approach has run-time complexity O(N^²). This complexity issue is usually mitigated with poor, but computationally cheap, approximate centroid optimization methods. In this paper, we first propose a solid theory on the linearization of the triplet loss with the use of class centroids, where the main conclusion is that our new linear loss represents a tight upper-bound to the triplet loss. Furthermore, based on the theory above, we propose a training algorithm that no longer requires the centroid optimization step, which means that our approach is the first in the field with a guaranteed linear run-time complexity. We show that the training of deep distance metric learning methods using the proposed upper-bound is substantially faster than triplet-based methods, while producing competitive retrieval accuracy results on benchmark datasets (CUB-200-2011 and CAR196). |
Rights: | ©2019 IEEE |
DOI: | 10.1109/CVPR.2019.01065 |
Grant ID: | http://purl.org/au-research/grants/arc/DP180103232 |
Published version: | https://ieeexplore.ieee.org/xpl/conhome/8938205/proceeding |
Appears in Collections: | Aurora harvest 8 Computer Science publications |
Files in This Item:
There are no files associated with this item.
Items in DSpace are protected by copyright, with all rights reserved, unless otherwise indicated.