mlsanityData to 14 Oct 2026

Privacy-Preserving Federated Action Recognition via Differentially Private Selective Tuning and Efficient Communication

arXiv 2603.21305 ↗First posted 22 Mar 2026cs.CVcited by 0

Not on the map. Why?

Abstract (arXiv)

Federated video action recognition enables collaborative model training without sharing raw video data, yet remains vulnerable to two key challenges: textitmodel exposure and textitcommunication overhead. Gradients exchanged between clients and the server can leak private motion patterns, while full-model synchronization of high-dimensional video networks causes significant bandwidth and communication costs. To address these issues, we propose textitFederated Differential Privacy with Selective Tuning and Efficient Communication for Action Recognition, namely textitFedDP-STECAR. Our textitFedDP-STECAR framework selectively fine-tunes and perturbs only a small subset of task-relevant layers under Differential Privacy (DP), reducing the surface of information leakage while preserving temporal coherence in video features. By transmitting only the tuned layers during aggregation, communication traffic is reduced by over 99% compared to full-model updates. Experiments on the UCF-101 dataset using the MViT-B-16x4 transformer show that textitFedDP-STECAR achieves up to 70.2% higher accuracy under strict privacy (ε=0.65) in centralized settings and 48% faster training with 73.1% accuracy in federated setups, enabling scalable and privacy-preserving video action recognition. Code available at https://github.com/izakariyya/mvit-federated-videodp

Key facts. No citing papers in the map yet (data through 14 Oct 2026). Builds on 16 of its 21 references that are in the map. First posted on arXiv 22 Mar 2026.

Builds on

16 of 21 references.

References · 16

  1. 2016Communication-Efficient Learning of Deep Networks from Decentralized Datacited by 7,669
  2. 2018Federated Optimization in Heterogeneous Networkscited by 2,987
  3. 2016Deep Learning with Differential Privacycited by 2,728
  4. 2020Tackling the Objective Inconsistency Problem in Heterogeneous Federated Optimizationcited by 906
  5. 2021Multiscale Vision Transformerscited by 897
  6. 2021Federated Learning Based on Dynamic Regularizationcited by 562
  7. 2019Federated Learning with Personalization Layerscited by 541
  8. 2017Differentially Private Federated Learning: A Client Level Perspectivecited by 459
  9. 2018Privacy Amplification by Subsampling: Tight Analyses via Couplings and Divergencescited by 242
  10. 2019Differentially Private Learning with Adaptive Clippingcited by 220
  11. 2021Gradient-Leakage Resilient Federated Learningcited by 40
  12. 2023Differentially Private Video Activity Recognitioncited by 9
4 more
  1. 2024Benchmarking Robust Self-Supervised Learning Across Diverse Downstream Taskscited by 5
  2. 2022Towards Efficient Active Learning of PDFAcited by 3
  3. 2024Activity Recognition on Avatar-Anonymized Datasets with Masked Differential Privacycited by 3
  4. 2025Video-DPRP: A Differentially Private Approach for Visual Privacy-Preserving Video Human Activity Recognitioncited by 3

Built on by

Not cited yet.