A Discriminatively Learned CNN Embedding for Person Re-identification

TL;DR

This paper introduces a novel siamese CNN architecture that combines identification and verification losses to learn more discriminative embeddings for person re-identification, achieving state-of-the-art results.

Contribution

It proposes a new CNN model that jointly optimizes identification and verification tasks for improved person re-ID performance.

Findings

Achieved state-of-the-art results on two public re-ID benchmarks.

The combined loss approach enhances discriminative feature learning.

Applicable to image retrieval tasks.

Abstract

We revisit two popular convolutional neural networks (CNN) in person re-identification (re-ID), i.e, verification and classification models. The two models have their respective advantages and limitations due to different loss functions. In this paper, we shed light on how to combine the two models to learn more discriminative pedestrian descriptors. Specifically, we propose a new siamese network that simultaneously computes identification loss and verification loss. Given a pair of training images, the network predicts the identities of the two images and whether they belong to the same identity. Our network learns a discriminative embedding and a similarity measurement at the same time, thus making full usage of the annotations. Albeit simple, the learned embedding improves the state-of-the-art performance on two public person re-ID benchmarks. Further, we show our architecture can also be applied in image retrieval.