UnrealPerson: An Adaptive Pipeline towards Costless Person Re-identification
This work addresses the high cost of annotated data collection and domain transfer for person re-identification, offering a more cost-effective solution for researchers and practitioners in this field.
This paper introduces UnrealPerson, a pipeline that uses synthetic image data to reduce costs in person re-identification (ReID) training and deployment. With 3,000 IDs and 120,000 instances, their method achieves a 38.5% rank-1 accuracy when directly transferred to MSMT17, nearly doubling previous records using synthetic data and surpassing some real data direct transfer records.
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.