FedVSR: Towards Model-Agnostic Federated Learning in Video Super-Resolution

2025-03-17

AI Summary

To address privacy risks arising from centralized training in video super-resolution (VSR), this work pioneers the integration of federated learning (FL) into this low-level vision task, proposing a lightweight, stateless, and model-agnostic distributed VSR framework. To overcome poor reconstruction quality and high communication overhead inherent in existing FL methods for VSR, we introduce a novel federated VSR paradigm and design a lightweight loss term that jointly optimizes local model updates and global aggregation—thereby enhancing both reconstruction accuracy and communication efficiency while preserving data privacy. Extensive experiments on multiple benchmark datasets demonstrate an average PSNR improvement of 0.85 dB over generic FL baselines, validating the effectiveness of our approach. The source code is publicly available.

Technology Category

• Machine Learning: Distributed Machine Learning & Federated Learning
• Computer Vision: Learning & Optimization for CV
• Search and Optimization: Learning to Search

Application Category

• User Modeling, Personalization and Recommendation: Federated recommendation systems and personalization
• Search and Retrieval-Augmented AI: Web learning to rank, online learning, and counterfactual learning for ranking
• Systems and Infrastructure for Web, Mobile and WoT: Federated Web and WoT systems, including distributed, federated and edge-based data processing

Abstract

Video Super-Resolution (VSR) reconstructs high-resolution videos from low-resolution inputs to restore fine details and improve visual clarity. While deep learning-based VSR methods achieve impressive results, their centralized nature raises serious privacy concerns, particularly in applications with strict privacy requirements. Federated Learning (FL) offers an alternative approach, but existing FL methods struggle with low-level vision tasks, leading to suboptimal reconstructions. To address this, we propose FedVSR1, a novel, architecture-independent, and stateless FL framework for VSR. Our approach introduces a lightweight loss term that improves local optimization and guides global aggregation with minimal computational overhead. To the best of our knowledge, this is the first attempt at federated VSR. Extensive experiments show that FedVSR outperforms general FL methods by an average of 0.85 dB in PSNR, highlighting its effectiveness. The code is available at: https://github.com/alimd94/FedVSR

Problem

• Privacy concerns in centralized video super-resolution methods
• Suboptimal reconstructions in federated learning for low-level vision tasks
• Need for architecture-independent federated learning in video super-resolution

Innovation

• Federated Learning for Video Super-Resolution
• Architecture-independent, stateless FL framework
• Lightweight loss term for improved local optimization