An Approximate GEMM Unit for Energy-Efficient Object Detection.

Pilipovic, Ratko; Risojevic, Vladimir; Bozic, Janko; Bulic, Patricio; Lotric, Uros

Pilipovic, Ratko; Risojevic, Vladimir; Bozic, Janko; Bulic, Patricio; Lotric, Uros.

Afiliación

Pilipovic R; Faculty of Computer and Information Science, University of Ljubljana, 1000 Ljubljana, Slovenia.
Risojevic V; Faculty of Electrical Engineering, University of Banja Luka, 78000 Banja Luka, Bosnia and Herzegovina.
Bozic J; Biotechnical Faculty, University of Ljubljana, 1000 Ljubljana, Slovenia.
Bulic P; Faculty of Computer and Information Science, University of Ljubljana, 1000 Ljubljana, Slovenia.
Lotric U; Faculty of Computer and Information Science, University of Ljubljana, 1000 Ljubljana, Slovenia.

Sensors (Basel) ; 21(12)2021 Jun 18.

Article en En | MEDLINE | ID: mdl-34207295

Asunto(s)

Inteligencia Artificial; Redes Neurales de la Computación; Algoritmos; Animales; Almacenamiento y Recuperación de la Información

Palabras clave

GEMM; YOLOv4-tiny; approximate computing; approximate general matrix multiplication; approximate multipliers; convolutional neural networks; energy-efficient processing; honeybee detection; matrix core; object detection; tensor core

Texto completo

Añadir a Mi BVS

Imprimir

XML

PubMed Links

Buscar en Google

Texto completo: 1 Colección: 01-internacional Base de datos: MEDLINE Asunto principal: Inteligencia Artificial / Redes Neurales de la Computación Tipo de estudio: Diagnostic_studies Límite: Animals Idioma: En Revista: Sensors (Basel) Año: 2021 Tipo del documento: Article País de afiliación: Eslovenia

Texto completo

Añadir a Mi BVS

Imprimir

XML

PubMed Links

Buscar en Google