Cooperative Perception: A Resource-Efficient Framework for Multi-Drone 3D Scene Reconstruction Using Federated Diffusion and NeRF

TL;DR

This paper presents a resource-efficient multi-drone perception framework that combines federated diffusion models, lightweight semantic extraction, and local NeRF updates for real-time 3D scene reconstruction, enhancing cooperative understanding while preserving privacy.

Contribution

It introduces a novel federated diffusion-based framework for multi-drone 3D scene reconstruction with semantic-aware compression, addressing computational and communication constraints.

Findings

Effective multi-agent 3D scene synthesis demonstrated in simulations

Enhanced privacy and scalability through federated learning

Potential for real-world deployment on drone testbeds

Abstract

The proposal introduces an innovative drone swarm perception system that aims to solve problems related to computational limitations and low-bandwidth communication, and real-time scene reconstruction. The framework enables efficient multi-agent 3D/4D scene synthesis through federated learning of shared diffusion model and YOLOv12 lightweight semantic extraction and local NeRF updates while maintaining privacy and scalability. The framework redesigns generative diffusion models for joint scene reconstruction, and improves cooperative scene understanding, while adding semantic-aware compression protocols. The approach can be validated through simulations and potential real-world deployment on drone testbeds, positioning it as a disruptive advancement in multi-agent AI for autonomous systems.