Frugal Federated Learning for Violence Detection: A Comparison of LoRA-Tuned VLMs and Personalized CNNs

Fuente: arXiv
Salvato in:
Dettagli Bibliografici
Autori principali: Thuau, Sébastien, Haidar, Siba, Bajracharya, Ayush, Chelouah, Rachid
Natura: Preprint
Pubblicazione: 2025
Soggetti:
Accesso online:
Tags: Aggiungi Tag
Nessun Tag, puoi essere il primo ad aggiungerne!!
_version_ 1866911221974827008
author Thuau, Sébastien
Haidar, Siba
Bajracharya, Ayush
Chelouah, Rachid
author_facet Thuau, Sébastien
Haidar, Siba
Bajracharya, Ayush
Chelouah, Rachid
contents We examine frugal federated learning approaches to violence detection by comparing two complementary strategies: (i) zero-shot and federated fine-tuning of vision-language models (VLMs), and (ii) personalized training of a compact 3D convolutional neural network (CNN3D). Using LLaVA-7B and a 65.8M parameter CNN3D as representative cases, we evaluate accuracy, calibration, and energy usage under realistic non-IID settings. Both approaches exceed 90% accuracy. CNN3D slightly outperforms Low-Rank Adaptation(LoRA)-tuned VLMs in ROC AUC and log loss, while using less energy. VLMs remain favorable for contextual reasoning and multimodal inference. We quantify energy and CO$_2$ emissions across training and inference, and analyze sustainability trade-offs for deployment. To our knowledge, this is the first comparative study of LoRA-tuned vision-language models and personalized CNNs for federated violence detection, with an emphasis on energy efficiency and environmental metrics. These findings support a hybrid model: lightweight CNNs for routine classification, with selective VLM activation for complex or descriptive scenarios. The resulting framework offers a reproducible baseline for responsible, resource-aware AI in video surveillance, with extensions toward real-time, multimodal, and lifecycle-aware systems.
format Preprint
id arxiv_https___arxiv_org_abs_2510_17651
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle Frugal Federated Learning for Violence Detection: A Comparison of LoRA-Tuned VLMs and Personalized CNNs
Thuau, Sébastien
Haidar, Siba
Bajracharya, Ayush
Chelouah, Rachid
Computer Vision and Pattern Recognition
Artificial Intelligence
Machine Learning
We examine frugal federated learning approaches to violence detection by comparing two complementary strategies: (i) zero-shot and federated fine-tuning of vision-language models (VLMs), and (ii) personalized training of a compact 3D convolutional neural network (CNN3D). Using LLaVA-7B and a 65.8M parameter CNN3D as representative cases, we evaluate accuracy, calibration, and energy usage under realistic non-IID settings. Both approaches exceed 90% accuracy. CNN3D slightly outperforms Low-Rank Adaptation(LoRA)-tuned VLMs in ROC AUC and log loss, while using less energy. VLMs remain favorable for contextual reasoning and multimodal inference. We quantify energy and CO$_2$ emissions across training and inference, and analyze sustainability trade-offs for deployment. To our knowledge, this is the first comparative study of LoRA-tuned vision-language models and personalized CNNs for federated violence detection, with an emphasis on energy efficiency and environmental metrics. These findings support a hybrid model: lightweight CNNs for routine classification, with selective VLM activation for complex or descriptive scenarios. The resulting framework offers a reproducible baseline for responsible, resource-aware AI in video surveillance, with extensions toward real-time, multimodal, and lifecycle-aware systems.
title Frugal Federated Learning for Violence Detection: A Comparison of LoRA-Tuned VLMs and Personalized CNNs
topic Computer Vision and Pattern Recognition
Artificial Intelligence
Machine Learning
url https://arxiv.org/abs/2510.17651