Pseudo-positive regularization for deep person re-identification

TL;DR

This paper introduces Pseudo Positive Regularization, a data augmentation technique that uses unlabeled external data to improve deep CNN training for person re-identification, reducing overfitting and enhancing performance.

Contribution

It proposes a novel Pseudo Positive Regularization method that enriches training data with unlabeled samples to improve CNN-based person re-ID accuracy.

Findings

Consistent performance improvement on CUHK03 and Market-1501 datasets.

Competitive results compared to state-of-the-art methods.

Effective data augmentation reduces overfitting.

Abstract

An intrinsic challenge of person re-identification (re-ID) is the annotation difficulty. This typically means 1) few training samples per identity, and 2) thus the lack of diversity among the training samples. Consequently, we face high risk of over-fitting when training the convolutional neural network (CNN), a state-of-the-art method in person re-ID. To reduce the risk of over-fitting, this paper proposes a Pseudo Positive Regularization (PPR) method to enrich the diversity of the training data. Specifically, unlabeled data from an independent pedestrian database is retrieved using the target training data as query. A small proportion of these retrieved samples are randomly selected as the Pseudo Positive samples and added to the target training set for the supervised CNN training. The addition of Pseudo Positive samples is therefore a data augmentation method to reduce the risk of over-fitting during CNN training. We implement our idea in the identification CNN models (i.e., CaffeNet, VGGNet-16 and ResNet-50). On CUHK03 and Market-1501 datasets, experimental results demonstrate that the proposed method consistently improves the baseline and yields competitive performance to the state-of-the-art person re-ID methods.