no code implementations • 6 Jul 2022 • Esteve Almirall, Davide Callegaro, Peter Bruins, Mar Santamaría, Pablo Martínez, Ulises Cortés
However, Digital Twins are data intensive and need highly localized data, making them difficult to scale, particularly to small cities, and with the high cost associated to data collection.
no code implementations • 11 Jan 2022 • Davide Callegaro, Francesco Restuccia, Marco Levorato
We extensively evaluate SmartDet on a real-world testbed composed of a JetSon Nano as mobile device and a GTX 980 Ti as edge server, connected through a Wi-Fi link.
2 code implementations • 7 Jan 2022 • Yoshitomo Matsubara, Davide Callegaro, Sameer Singh, Marco Levorato, Francesco Restuccia
We show that BottleFit decreases power consumption and latency respectively by up to 49% and 89% with respect to (w. r. t.)
2 code implementations • 20 Nov 2020 • Yoshitomo Matsubara, Davide Callegaro, Sabur Baidya, Marco Levorato, Sameer Singh
In this paper, we propose to modify the structure and training process of DNN models for complex image classification tasks to achieve in-network compression in the early network layers.
2 code implementations • 1 Oct 2019 • Yoshitomo Matsubara, Sabur Baidya, Davide Callegaro, Marco Levorato, Sameer Singh
Offloading the execution of complex Deep Neural Networks (DNNs) models to compute-capable devices at the network edge, that is, edge servers, can significantly reduce capture-to-output delay.