Depth estimation is crucial in several computer vision applications, and a recent trend aims at inferring such a cue from a single camera through computationally demanding CNNs - precluding their practical deployment in several application contexts characterized by low-power constraints. Purposely, we develop a tiny network tailored to microcontrollers, processing low-resolution images to obtain a coarse depth map of the observed scene. Our solution enables depth perception with minimal power requirements (a few hundreds of mW), accurately enough to pave the way to several high-level applications at-the-edge.
Enabling monocular depth perception at the very edge / Peluso V.; Cipolletta A.; Calimera A.; Poggi M.; Tosi F.; Aleotti F.; Mattoccia S.. - ELETTRONICO. - (2020), pp. 9150972.1581-9150972.1583. (Intervento presentato al convegno 2020 IEEE/CVF Conference on Computer Vision and Pattern Recognition Workshops, CVPRW 2020 tenutosi a Seattle, WA, USA nel 14-19 June 2020) [10.1109/CVPRW50498.2020.00204].
Enabling monocular depth perception at the very edge
Poggi M.;Tosi F.;Aleotti F.;Mattoccia S.
2020
Abstract
Depth estimation is crucial in several computer vision applications, and a recent trend aims at inferring such a cue from a single camera through computationally demanding CNNs - precluding their practical deployment in several application contexts characterized by low-power constraints. Purposely, we develop a tiny network tailored to microcontrollers, processing low-resolution images to obtain a coarse depth map of the observed scene. Our solution enables depth perception with minimal power requirements (a few hundreds of mW), accurately enough to pave the way to several high-level applications at-the-edge.File | Dimensione | Formato | |
---|---|---|---|
Peluso_Enabling_Monocular_Depth_Perception_at_the_Very_Edge_CVPRW_2020_paper.pdf
accesso aperto
Tipo:
Postprint
Licenza:
Licenza per accesso libero gratuito
Dimensione
572.06 kB
Formato
Adobe PDF
|
572.06 kB | Adobe PDF | Visualizza/Apri |
I documenti in IRIS sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.