Deep Cosine Metric Learning for Person Re-identification
Authors:
- Nicolai WojkeNWNicolai Wojke
- Alex BewleyABAlex Bewley
- 1 March 2018
- conference paper
- Published by Institute of Electrical and Electronics Engineers (IEEE)
- p. 748-756
Abstract
Metric learning aims to construct an embedding where two extracted features corresponding to the same identity are likely to be closer than features from different identities. This paper presents a method for learning such a feature space where the cosine similarity is effectively optimized through a simple re-parametrization of the conventional softmax classification regime. At test time, the final classification layer can be stripped from the network to facilitate nearest neighbor queries on unseen individuals using the cosine similarity metric. This approach presents a simple alternative to direct metric learning objectives such as siamese networks that have required sophisticated pair or triplet sampling strategies in the past. The method is evaluated on two large-scale pedestrian re-identification datasets where competitive results are achieved overall. In particular, we achieve better generalization on the test set compared to a network trained with triplet loss.Keywords
All Related Versions
This publication has 24 references indexed in Scilit:
- Deeply-Learned Part-Aligned Representations for Person Re-identification2017
- Simple online and realtime tracking with a deep association metric2017
- Person Re-Identification by Deep Joint Learning of Multi-Loss Classification2017
- See the Forest for the Trees: Joint Spatial and Temporal Recurrent Neural Networks for Video-Based Person Re-identification2017
- Learning Deep Context-Aware Features over Body and Latent Parts for Person Re-identification2017
- Person Re-identification in the Wild2017
- MARS: A Video Benchmark for Large-Scale Person Re-Identification2016
- Scalable Person Re-identification: A Benchmark2015
- DeepReID: Deep Filter Pairing Neural Network for Person Re-identification2014
- Large scale metric learning from equivalence constraints2012