UnrealPerson: An Adaptive Pipeline towards Costless Person Re-identification

TL;DR

UnrealPerson introduces a cost-effective pipeline utilizing high-quality synthesized unreal images with controllable distributions and annotations to improve person re-identification accuracy and adaptability across domains.

Contribution

The paper presents a novel UnrealPerson pipeline that leverages synthesized high-quality images with controllable distributions and annotations for cost-effective person ReID.

Findings

Achieves 38.5% rank-1 accuracy on MSMT17 with synthesized data.

Nearly doubles previous results using synthesized data.

Surpasses prior direct transfer results with real data.

Abstract

The main difficulty of person re-identification (ReID) lies in collecting annotated data and transferring the model across different domains. This paper presents UnrealPerson, a novel pipeline that makes full use of unreal image data to decrease the costs in both the training and deployment stages. Its fundamental part is a system that can generate synthesized images of high-quality and from controllable distributions. Instance-level annotation goes with the synthesized data and is almost free. We point out some details in image synthesis that largely impact the data quality. With 3,000 IDs and 120,000 instances, our method achieves a 38.5% rank-1 accuracy when being directly transferred to MSMT17. It almost doubles the former record using synthesized data and even surpasses previous direct transfer records using real data. This offers a good basis for unsupervised domain adaption, where our pre-trained model is easily plugged into the state-of-the-art algorithms towards higher accuracy. In addition, the data distribution can be flexibly adjusted to fit some corner ReID scenarios, which widens the application of our pipeline. We will publish our data synthesis toolkit and synthesized data in https://github.com/FlyHighest/UnrealPerson.