Local-Adaptive Face Recognition via Graph-based Meta-Clustering and Regularized Adaptation
This addresses privacy-preserving continuous learning for face recognition in specific environments, though it appears incremental as it builds on existing federated and unsupervised adaptation concepts.
The paper tackles the problem of adapting face recognition models to local environments without access to labeled data or centralized data transfer, introducing the Local-Adaptive Face Recognition (LaFR) setup. They achieve this via a graph-based meta-clustering method with regularization, showing effectiveness in racial and sensor adaptation experiments.
Due to the rising concern of data privacy, it's reasonable to assume the local client data can't be transferred to a centralized server, nor their associated identity label is provided. To support continuous learning and fill the last-mile quality gap, we introduce a new problem setup called Local-Adaptive Face Recognition (LaFR). Leveraging the environment-specific local data after the deployment of the initial global model, LaFR aims at getting optimal performance by training local-adapted models automatically and un-supervisely, as opposed to fixing their initial global model. We achieve this by a newly proposed embedding cluster model based on Graph Convolution Network (GCN), which is trained via meta-optimization procedure. Compared with previous works, our meta-clustering model can generalize well in unseen local environments. With the pseudo identity labels from the clustering results, we further introduce novel regularization techniques to improve the model adaptation performance. Extensive experiments on racial and internal sensor adaptation demonstrate that our proposed solution is more effective for adapting face recognition models in each specific environment. Meanwhile, we show that LaFR can further improve the global model by a simple federated aggregation over the updated local models.