In Defense of the Triplet Loss for Person Re-Identification

TL;DR

This paper demonstrates that the triplet loss, when used for end-to-end deep metric learning, outperforms most other methods in person re-identification, challenging the belief that it is inferior to surrogate losses.

Contribution

The authors show that a variant of the triplet loss is highly effective for end-to-end deep metric learning in person re-identification, outperforming other approaches.

Findings

Triplet loss outperforms surrogate losses in person re-identification.

End-to-end training with triplet loss yields superior results.

Model trained from scratch or pretrained benefits from triplet loss.

Abstract

In the past few years, the field of computer vision has gone through a revolution fueled mainly by the advent of large datasets and the adoption of deep convolutional neural networks for end-to-end learning. The person re-identification subfield is no exception to this. Unfortunately, a prevailing belief in the community seems to be that the triplet loss is inferior to using surrogate losses (classification, verification) followed by a separate metric learning step. We show that, for models trained from scratch as well as pretrained ones, using a variant of the triplet loss to perform end-to-end deep metric learning outperforms most other published methods by a large margin.