FLcode-lab/DADPFed-Code: DADPFed v1.0.0: Code for "Drift-Aware Dynamic Pruning: A Stabilising Heuristic for Heterogeneous Federated Action Recognition"

Read full text Summarise with VIRAT
Authors: FLcode-lab
Cited by: 0

Summary

No summary yet. VIRAT can read this paper and write a plain-language summary of its aim, methods, key findings and limitations.

Summarise with VIRAT Free to try with your Science Catalogs account.

Abstract

This is the official code release accompanying the manuscript submitted to The Visual Computer. Repository Overview Paper Title: Drift-Aware Dynamic Pruning: A Stabilising Heuristic for Heterogeneous Federated Action Recognition Journal: The Visual Computer Authors: Zhihao Liu, Wei Guo, Jie Wu, Mengke Zhu, Jiamin Liang Contents Full implementation of DADPFed framework Training/evaluation scripts for federated action recognition Subset construction, split files, and preprocessing pipelines for martial-arts video benchmarks Detailed README with environment setup and reproduction commands Citation If you use this code in your research, please cite our manuscript: Zhihao Liu, Wei Guo, Jie Wu, Mengke Zhu, Jiamin Liang. Drift-Aware Dynamic Pruning: A Stabilising Heuristic for Heterogeneous Federated Action Recognition. The Visual Computer, 2025. Data Note Raw video data is derived from public benchmarks (Kinetics-MA, UCF101-MA). Due to copyright constraints, raw videos are not included; all subset construction and preprocessing scripts are provided for full reproducibility.

Fields of Study & Subjects

Human Pose and Action Recognition Privacy-Preserving Technologies in Data Hand Gesture Recognition Systems

Publication details

Journal / source
Zenodo (CERN European Organization for Nuclear Research)
Publication date
30 March 2026
Language
EN