Automatic classification of video using a scalable photonic neuro-inspired architecture - Archive ouverte HAL Accéder directement au contenu
Communication Dans Un Congrès Année : 2020

Automatic classification of video using a scalable photonic neuro-inspired architecture

Résumé

We propose a physical alternative of software based approaches for advanced classification task by considering a photonic-based architecture implementing a recurrent neural network with up to 16,384 physical neurons. This architecture is realized with o↵-the-shelf components and can be scaled up to hundred thousand or millions of nodes while ensuring data-ecient training strategy thanks to the reservoir computing framework. We use this architecture to perform a challenging computer vision task: the classification of human actions from a video feed. For this task, we show for the first time that a physical architecture with a simple learning strategy, consisting of training one linear readout for each class, can achieve a >90% success rate in terms of classification accuracy. This rivals the deep-learning approaches in terms of level of performance and hence could pave the way towards novel paradigm for ecient real-time video processing at the physical layer using photonic systems.
Fichier non déposé

Dates et versions

hal-03422999 , version 1 (09-11-2021)

Identifiants

  • HAL Id : hal-03422999 , version 1

Citer

Damien Rontani, Piotr Antonik, Nicolas Marsal, Daniel Brunner. Automatic classification of video using a scalable photonic neuro-inspired architecture. SPIE Photonics Europe, Mar 2020, Strasbourg, France. ⟨hal-03422999⟩
4 Consultations
0 Téléchargements

Partager

Gmail Facebook X LinkedIn More