Identity Preserving Generative Adversarial Network for Cross-Domain Person Re-identification
This addresses the scalability and usability issue in real-world person re-identification by improving cross-domain generalization, though it is incremental as it builds on existing GAN and domain adaptation methods.
The paper tackles the problem of person re-identification models failing to generalize from a labeled source domain to an unlabeled target domain due to domain bias, proposing an identity-preserving generative adversarial network (IPGAN) to translate images across domains while preserving identity, and achieving competitive accuracy on benchmarks like Market-1501 and DukeMTMC-reID.
Person re-identification is to retrieval pedestrian images from no-overlap camera views detected by pedestrian detectors. Most existing person re-identification (re-ID) models often fail to generalize well from the source domain where the models are trained to a new target domain without labels, because of the bias between the source and target domain. This issue significantly limits the scalability and usability of the models in the real world. Providing a labeled source training set and an unlabeled target training set, the aim of this paper is to improve the generalization ability of re-ID models to the target domain. To this end, we propose an image generative network named identity preserving generative adversarial network (IPGAN). The proposed method has two excellent properties: 1) only a single model is employed to translate the labeled images from the source domain to the target camera domains in an unsupervised manner; 2) The identity information of images from the source domain is preserved before and after translation. Furthermore, we propose IBN-reID model for the person re-identification task. It has better generalization ability than baseline models, especially in the cases without any domain adaptation. The IBN-reID model is trained on the translated images by supervised methods. Experimental results on Market-1501 and DukeMTMC-reID show that the images generated by IPGAN are more suitable for cross-domain person re-identification. Very competitive re-ID accuracy is achieved by our method.