Energy-efficient use of an embedded heterogeneous SoC for the inference of CNNs

Agathe Archet (1, 2) , Nicolas Ventroux (3) , Nicolas Gac (2) , François Orieux (2)

1 Thales Research and Technology [Palaiseau]
2 L2S - Laboratoire des signaux et systèmes
3 TRT - Thales Research and Technologies [Orsay]

Mots clés: Heterogeneous computing; Convolution Neural Networks; Low-power inference; DLA; Computer vision; Jetson AGX Orin

Domaines: Systèmes embarqués; Intelligence artificielle [cs.AI]; Vision par ordinateur et reconnaissance de formes [cs.CV]

Agathe Archet, Nicolas Ventroux, Nicolas Gac, François Orieux. Energy-efficient use of an embedded heterogeneous SoC for the inference of CNNs. 2023 26th Euromicro Conference on Digital System Design (DSD), Sep 2023, Durrës, Albania. ⟨hal-04148582⟩