Personalized Federated Learning for Egocentric Video Gaze Estimation with Comprehensive Parameter Frezzing
This addresses the problem of adapting gaze estimation models to individual users in federated settings, but it is incremental as it builds on existing PFL methods with a specific freezing strategy.
The paper tackled personalized federated learning for egocentric video gaze estimation by freezing significant parameters in client models, achieving superior recall, precision, and F1-score on EGTEA Gaze+ and Ego4D datasets.
Egocentric video gaze estimation requires models to capture individual gaze patterns while adapting to diverse user data. Our approach leverages a transformer-based architecture, integrating it into a PFL framework where only the most significant parameters, those exhibiting the highest rate of change during training, are selected and frozen for personalization in client models. Through extensive experimentation on the EGTEA Gaze+ and Ego4D datasets, we demonstrate that FedCPF significantly outperforms previously reported federated learning methods, achieving superior recall, precision, and F1-score. These results confirm the effectiveness of our comprehensive parameters freezing strategy in enhancing model personalization, making FedCPF a promising approach for tasks requiring both adaptability and accuracy in federated learning settings.